[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-firmai--business-machine-learning":3,"tool-firmai--business-machine-learning":61},[4,18,26,36,44,53],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",140436,2,"2026-04-05T23:32:43",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":10,"last_commit_at":50,"category_tags":51,"status":17},4292,"Deep-Live-Cam","hacksider\u002FDeep-Live-Cam","Deep-Live-Cam 是一款专注于实时换脸与视频生成的开源工具，用户仅需一张静态照片，即可通过“一键操作”实现摄像头画面的即时变脸或制作深度伪造视频。它有效解决了传统换脸技术流程繁琐、对硬件配置要求极高以及难以实时预览的痛点，让高质量的数字内容创作变得触手可及。\n\n这款工具不仅适合开发者和技术研究人员探索算法边界，更因其极简的操作逻辑（仅需三步：选脸、选摄像头、启动），广泛适用于普通用户、内容创作者、设计师及直播主播。无论是为了动画角色定制、服装展示模特替换，还是制作趣味短视频和直播互动，Deep-Live-Cam 都能提供流畅的支持。\n\n其核心技术亮点在于强大的实时处理能力，支持口型遮罩（Mouth Mask）以保留使用者原始的嘴部动作，确保表情自然精准；同时具备“人脸映射”功能，可同时对画面中的多个主体应用不同面孔。此外，项目内置了严格的内容安全过滤机制，自动拦截涉及裸露、暴力等不当素材，并倡导用户在获得授权及明确标注的前提下合规使用，体现了技术发展与伦理责任的平衡。",88924,"2026-04-06T03:28:53",[14,15,13,52],"视频",{"id":54,"name":55,"github_repo":56,"description_zh":57,"stars":58,"difficulty_score":32,"last_commit_at":59,"category_tags":60,"status":17},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[14,35],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":66,"readme_en":67,"readme_zh":68,"quickstart_zh":69,"use_case_zh":70,"hero_image_url":71,"owner_login":72,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":76,"owner_location":75,"owner_email":77,"owner_twitter":78,"owner_website":79,"owner_url":80,"languages":81,"stars":86,"forks":87,"last_commit_at":88,"license":89,"difficulty_score":90,"env_os":91,"env_gpu":92,"env_ram":92,"env_deps":93,"category_tags":96,"github_topics":98,"view_count":32,"oss_zip_url":75,"oss_zip_packed_at":75,"status":17,"created_at":105,"updated_at":106,"faqs":107,"releases":108},4322,"firmai\u002Fbusiness-machine-learning","business-machine-learning","A curated list of practical business machine learning (BML) and business data science (BDS) applications for Accounting, Customer, Employee, Legal, Management and Operations (by @firmai)","business-machine-learning 是一份由社区精心整理的实用资源清单，专注于将机器学习与数据科学应用于真实的商业场景。它系统地涵盖了会计、客户管理、人力资源、法律合规、企业运营及管理层决策等核心领域，旨在填补前沿算法与具体业务需求之间的鸿沟。\n\n面对企业在数字化转型中“有数据却难落地”的痛点，这份清单提供了从客户生命周期预测、流失预警、员工绩效分析到非传统数据（如卫星图像、社交媒体情绪）挖掘等一系列经过验证的应用案例与方法论。它不仅罗列了技术工具，还整合了相关的研究文章、课程资源及数据分析框架，帮助用户快速找到解决特定业务问题的最佳实践。\n\n该资源特别适合从事商业数据分析的研究人员、希望将 AI 技术落地的企业开发者、数据科学家以及寻求创新解决方案的管理顾问。无论是需要构建预测模型的技术专家，还是希望了解行业趋势的决策者，都能从中获得极具价值的参考。其独特亮点在于不仅关注传统的结构化数据，还积极探索 GitHub 活动日志、网络流量等另类数据在金融与投资洞察中的应用，为量化分析和战略规划开辟了新的视角。","# Business Machine Learning and Data Science Applications \n\n---------\n\n\n## 🌟 We Are Growing!\n\nWe're seeking to collaborate with motivated, independent PhD graduates or doctoral students on approximately seven new projects in 2024. If you’re interested in contributing to cutting-edge investment insights and data analysis, please get in touch! This could be in colaboration with a university or as independent study. \n\n![image](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ffirmai_business-machine-learning_readme_dbe070c495d5.png)\n\n\n### 🚀 About Sov.ai\n\nSov.ai is at the forefront of integrating advanced machine learning techniques with financial data analysis to revolutionize investment strategies. We are working with **three of the top 10** quantitative hedge funds, and with many mid-sized and boutique firms. \n\nOur platform leverages diverse data sources and innovative algorithms to deliver actionable insights that drive smarter investment decisions. \n\nBy joining Sov.ai, you'll be part of a dynamic research team dedicated to pushing the boundaries of what's possible in finance through technology. Before expressing your interest, please be aware that the research will be predominantly challenging and experimental in nature.\n\n\n### 🔍 Research and Project Opportunities\n\nWe offer a wide range of projects that cater to various interests and expertise within machine learning and finance. Some of the exciting recent projects include:\n\n- **Predictive Modeling with GitHub Logs:** Develop models to predict market trends and investment opportunities using GitHub activity and developer data.\n- **Satallite Data Analysis:** Explore non-traditional data sources such as social media sentiment, satellite imagery, or web traffic to enhance financial forecasting.\n- **Data Imputation Techniques:** Investigate new methods for handling missing or incomplete data to improve the robustness and accuracy of our models.\n\nPlease visit [docs.sov.ai](https:\u002F\u002Fdocs.sov.ai) for more information on public projects that have made it into the subscription product. If you already have a corporate sponsor, we are also happy to work with them. \n\n### 🌐 Why Join Sov.ai?\n\n- **Innovative Environment:** Engage with the latest technologies and methodologies in machine learning and finance.\n- **Collaborative Team:** Work alongside a team of experts passionate about driving innovation in investment insights.\n- **Flexible Projects:** Tailor your research to align with your interests and expertise, with the freedom to explore new ideas.\n- **Experienced Researchers:** Experts previously from NYU, Columbia, Oxford-Man Institute, Alan Turing Institute, and Cambridge.\n- **Post Research:** Connect with alumni that has moved on to DRW, Citadel Securities, Virtu Financial, Akuna Capital, HRT.\n\n\n### 🤝 How to Apply\n\nIf you’re excited about leveraging your expertise in machine learning and finance to drive impactful research and projects, we’d love to hear from you! Please reach out to us at [research@sov.ai](mailto:research@sov.ai) with your resume and a brief description of your research interests.\n\nJoin us in shaping the future of investment insights and making a meaningful impact in the world of finance!\n\n\n\n\n## Table of Contents\n\n### Department Applications\n\u003C!-- MarkdownTOC depth=4 -->\n\n- [Accounting](#accounting)\n    - [Machine Learning](#accounting-ml)\n    - [Analytics](#accounting-analytics)\n    - [Textual Analysis](#accounting-text)\n    - [Data](#accounting-data)\n    - [Research and Articles](#accounting-ra)\n    - [Websites](#accounting-web)\n    - [Courses](#accounting-course)\n- [Customer](#customer)\n    - [Lifetime Value](#customer-clv)\n    - [Segmentation](#customer-seg)\n    - [Behaviour](#customer-behave)\n    - [Recommender](#customer-rec)\n    - [Churn Prediction](#customer-cp)\n    - [Sentiment](#customer-sent)\n- [Employee](#employee)\n    - [Management](#employee-man)\n    - [Performance](#employee-perf)\n    - [Turnover](#employee-general-turn)\n    - [Conversations](#employee-con)\n    - [Physical](#employee-ph)\n- [Legal](#legal)\n    - [Tools](#legal-tools)\n    - [Policy and Regulatory](#legal-pr)\n    - [Judicial](#legal-judicial)\n- [Management](#management)\n    - [Strategy](#management-strat)\n    - [Decision Optimisation](#management-do)\n    - [Causal Inference](#management-causal)\n    - [Statistics](#management-stat)\n    - [Quantitative](#management-quant)\n    - [Data](#management-data)\n- [Operations](#operations)\n    - [Failures and Anomalies](#operations-fail)\n    - [Load and Capacity Management](#operations-load)\n    - [Prediction Management](#operations-predict)\n\n\n\u003C!-- \u002FMarkdownTOC -->\n#### Also see [Python Business Analytics](https:\u002F\u002Fgithub.com\u002Ffirmai\u002Fpython-business-analytics)\n\n\u003Ca name=\"accounting\">\u003C\u002Fa>\n## Accounting\n\n\u003Ca name=\"accounting-ml\">\u003C\u002Fa>\n#### Machine Learning\n* [Chart of Account Prediction](https:\u002F\u002Fgithub.com\u002Fagdgovsg\u002Fml-coa-charging ) - Using labeled data to suggest the account name for every transaction.\n* [Accounting Anomalies](https:\u002F\u002Fgithub.com\u002FGitiHubi\u002FdeepAI\u002Fblob\u002Fmaster\u002FGTC_2018_Lab-solutions.ipynb) -  Using deep-learning frameworks to identify accounting anomalies.\n* [Financial Statement Anomalies](https:\u002F\u002Fgithub.com\u002Frameshcalamur\u002Ffin-stmt-anom) - Detecting anomalies before filing, using R.\n* [Useful Life Prediction (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FAged%20Debtors\u002F) - Predict the useful life of assets using sensor observations and feature engineering.\n* [AI Applied to XBRL](https:\u002F\u002Fgithub.com\u002FNiels-Peter\u002FXBRL-AI) - Standardized representation of XBRL into AI and Machine learning.\n \n\u003Ca name=\"accounting-analytics\">\u003C\u002Fa>\n#### Analytics\n\n* [Forensic Accounting](https:\u002F\u002Fgithub.com\u002Fmschermann\u002Fforensic_accounting) - Collection of case studies on forensic accounting using data analysis.  On the lookout for more data to practise forensic accounting, *please get in [touch](https:\u002F\u002Fgithub.com\u002Fmschermann\u002F)* \n* [General Ledger (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FGeneral%20Ledger\u002F) - Data processing over a general ledger as exported through an accounting system.\n* [Bullet Graph (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FBullet-Graph-Article\u002F) - Bullet graph visualisation helpful for tracking sales, commission and other performance.\n* [Aged Debtors (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FAged%20Debtors\u002F) - Example analysis to invetigate aged debtors.\n* [Automated FS XBRL](https:\u002F\u002Fgithub.com\u002FCharlesHoffmanCPA\u002FcharleshoffmanCPA.github.io) - XML Language, however, possibly port analysis into Python.\n\n\u003Ca name=\"accounting-text\">\u003C\u002Fa>\n#### Textual Analysis\n\n* [Financial Sentiment Analysis](https:\u002F\u002Fgithub.com\u002FEricHe98\u002FFinancial-Statements-Text-Analysis) - Sentiment, distance and proportion analysis for trading signals.\n* [Extensive NLP](https:\u002F\u002Fgithub.com\u002FTiesdeKok\u002FPython_NLP_Tutorial\u002Fblob\u002Fmaster\u002FNLP_Notebook.ipynb) - Comprehensive NLP techniques for accounting research.\n\n\u003Ca name=\"accounting-data\">\u003C\u002Fa>\n#### Data, Parsing and APIs\n\n* [EDGAR](https:\u002F\u002Fgithub.com\u002FTiesdeKok\u002FUW_Python_Camp\u002Fblob\u002Fmaster\u002FMaterials\u002FSession_5\u002FEDGAR_walkthrough.ipynb) - A walk-through in how to obtain EDGAR data. \n* [IRS](http:\u002F\u002Fsocial-metrics.org\u002Fsox\u002F) - Acessing and parsing IRS filings.\n* [Financial Corporate](http:\u002F\u002Fraw.rutgers.edu\u002FCorporate%20Financial%20Data.html) - Rutgers corporate financial datasets.\n* [Non-financial Corporate](http:\u002F\u002Fraw.rutgers.edu\u002FNon-Financial%20Corporate%20Data.html) - Rutgers non-financial corporate dataset.\n* [PDF Parsing](https:\u002F\u002Fgithub.com\u002Fdanshorstein\u002Fpython4cpas\u002Fblob\u002Fmaster\u002F03_parsing_pdf_files\u002FAR%20Aging%20-%20working.ipynb) - Extracting useful data from PDF documents. \n* [PDF Tabel to Excel](https:\u002F\u002Fgithub.com\u002Fdanshorstein\u002Fficpa_article) - How to output an excel file from a PDF.\n\n\u003Ca name=\"accounting-ra\">\u003C\u002Fa>\n#### Research And Articles\n\n* [Understanding Accounting Analytics](http:\u002F\u002Fsocial-metrics.org\u002Faccountinganalytics\u002F) - An article that tackles the importance of accounting analytics.\n* [VLFeat](http:\u002F\u002Fwww.vlfeat.org\u002F) - VLFeat is an open and portable library of computer vision algorithms, which has Matlab toolbox.\n\n\u003Ca name=\"accounting-web\">\u003C\u002Fa>\n#### Websites\n\n* [Rutgers Raw](http:\u002F\u002Fraw.rutgers.edu\u002F) - Good digital accounting research from Rutgers.\n\n\u003Ca name=\"accounting-course\">\u003C\u002Fa>\n#### Courses\n\n* [Computer Augmented Accounting](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLauepKFT6DK8TaNaq_SqZW4LIDJhCkZe2) - A video series from Rutgers University looking at the use of computation to improve accounting.\n* [Accounting in a Digital Era](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLauepKFT6DK8_Xun584UQPPsg1grYkWw0) - Another series by Rutgers investigating the effects the digital age will have on accounting.\n\n\u003Ca name=\"customer\">\u003C\u002Fa>\n## Customer\n\n\u003Ca name=\"customer-clv\">\u003C\u002Fa>\n#### Lifetime Value\n* [Pareto\u002FNBD Model](https:\u002F\u002Fgithub.com\u002Fzabahana\u002FCustomer-LifeTime-Value-Analysis\u002Fblob\u002Fmaster\u002FCLV%20Analysis.ipynb) - Calculate the CLV using a Pareto\u002FNBD model.\n* [Cohort Analysis](https:\u002F\u002Fgithub.com\u002Firis9112\u002FCustomer-Segmentation\u002Fblob\u002Fmaster\u002FChapter1-Cohort_Analysis.ipynb) - Cohort analysis to group customers into mutually exclusive cohorts measured over time. \n\n \n\u003Ca name=\"customer-seg\">\u003C\u002Fa>\n#### Segmentation\n\n* [E-commerce](https:\u002F\u002Fgithub.com\u002Fjalajthanaki\u002FCustomer_segmentation\u002Fblob\u002Fmaster\u002FCust_segmentation_online_retail.ipynb ) - E-commerce customer segmentation.\n* [Groceries](https:\u002F\u002Fgithub.com\u002Fharry329\u002FCustomerFinding\u002Fblob\u002Fmaster\u002Fcustomer_segments.ipynb ) - Segmentation for grocery customers. \n* [Online Retailer](https:\u002F\u002Fgithub.com\u002FVinayak02\u002FCustomerCentricRetail\u002Fblob\u002Fmaster\u002FCustomerSegmentation\u002FCustomer_Segmentation_Online_Retail.ipynb) - Online retailer segmentation.\n* [Bank](https:\u002F\u002Fgithub.com\u002FMogbo\u002FCustomer-Clustering-Segmentation-Challenge) - Bank customer segmentation.\n* [Wholesale](https:\u002F\u002Fgithub.com\u002FSyedAdilAli93\u002FIdentifying-Customers\u002Fblob\u002Fmaster\u002Fcustomer_segments.ipynb) - Clustering of wholesale customers.\n* [Various](https:\u002F\u002Fgithub.com\u002Fabalaji-blr\u002FCustomerSegments\u002Ftree\u002Fmaster\u002Fdeliver ) - Multiple types of segmentation and clustering techniques. \n\n\n\u003Ca name=\"customer-behave\">\u003C\u002Fa>\n#### Behaviour\n\n* [RNN](https:\u002F\u002Fgithub.com\u002FDaniSanchezSantolaya\u002FRNN-customer-behavior\u002Ftree\u002Fmaster\u002Fsrc) - Investigating customer behaviour over time with sequential analysis using an RNN model.\n* [Neural Net](https:\u002F\u002Fgithub.com\u002FVinayak02\u002FCustomerCentricRetail\u002Fblob\u002Fmaster\u002FDemandForecasting\u002FNeuralNetworks.ipynb) - Demand forecasting using artificial neural networks.\n* [Temporal Analytics](https:\u002F\u002Fgithub.com\u002Friccotti\u002FCustomerTemporalRegularities) - Investigating customer temporal regularities.\n* [POS Analytics](https:\u002F\u002Fgithub.com\u002FIBM\u002Fcustomer_pos_analytics\u002Fblob\u002Fmaster\u002Fcode\u002FCustomer%20Ranking%20POS%20wip.ipynb) - Analytics driven customer behaviour ranking for retail promotions using POS data.\n* [Wholesale Customer](https:\u002F\u002Fgithub.com\u002Fkralmachine\u002FWholesaleCustomerAnalysis\u002Fblob\u002Fmaster\u002FWhosaleCustomerAnalysis.ipynb) - Wholesale customer exploratory data analysis.\n* [RFM](https:\u002F\u002Fgithub.com\u002Fespin086\u002Fcustomer_growth\u002Fblob\u002Fmaster\u002Frfm\u002Frfm.ipynb) - Doing a RFM (recency, frequency, monetary) analysis. \n* [Returns Behaviour](https:\u002F\u002Fgithub.com\u002Fadarsh2111\u002FCustomer-Returns-Analysis-Customer-Fraud-Detection-\u002Fblob\u002Fmaster\u002FReturns%20Analysis.ipynb) - Predicting total returns and fraudulent returns. \n* [Visits](https:\u002F\u002Fgithub.com\u002FRyanfras\u002FCustomer-Visits\u002Fblob\u002Fmaster\u002FCustomer%20Visits.ipynb) - Predicting which day of week a customer will visit.\n* [Bank: Next Purchase](https:\u002F\u002Fgithub.com\u002Falbertcdc\u002FProject_CAJAMAR) - A project to predict bank customers' most probable next purchase.\n* [Bank: Customer Prediction](https:\u002F\u002Fgithub.com\u002Frohangawade\u002FPredicting-Target-customers-for-Bank-Policy-subscribtion-using-Logistic_Regression_Transparency) - Predicting Target customers who will subscribe the new policy of the bank.\n* [Next Purchase](https:\u002F\u002Fgithub.com\u002FFeaturetools\u002Fpredict-next-purchase) - Predict a customers’ next purchase also using feature engineering. \n* [Customer Purchase Repeats](https:\u002F\u002Fgithub.com\u002Fkpei\u002FCustomer-Analytics\u002Fblob\u002Fmaster\u002Fcustomer_zakka.ipynb) - Using the lifetimes python library and real jewellery retailer data analyse customer repeat purchases.\n* [AB Testing](https:\u002F\u002Fgithub.com\u002Fsushant2811\u002FcustomerAnalyticsWithA-BTesting\u002Fblob\u002Fmaster\u002FcustomerAnalyticsWithA-BTesting.ipynb) - Find the best KPI and do A\u002FB testing.\n* [Customer Survey (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FCustomer%20Survey\u002F) - Example of parsing and analysing a customer survey. \n* [Happiness](https:\u002F\u002Fgithub.com\u002Frohit6205\u002FpredictHappiness\u002Fblob\u002Fmaster\u002FpredictingHapiness.ipynb) - Analysing customer happiness from hotel stays using reviews. \n* [Miscellaneous Customer Analytics](https:\u002F\u002Fgithub.com\u002Fmapr-demos\u002Fcustomer360) - Various tools and techniques for customer analysis. \n\n\n\u003Ca name=\"customer-rec\">\u003C\u002Fa>\n#### Recommender\n\n* [Recommendation](https:\u002F\u002Fgithub.com\u002Fannalucia1\u002FCustomer-Behavior-Analysis-Recommendation\u002Fblob\u002Fmaster\u002Frecomendation_by_RatingScore.ipynb) - Recommend the songs that a customer on a music app would prefer listening to. \n* [General Recommender](https:\u002F\u002Fgithub.com\u002FVinayak02\u002FCustomerCentricRetail\u002Fblob\u002Fmaster\u002FRecommenderSystem\u002FRecommender.ipynb) - Identifying which products to recommend to which customers. \n* [Collaborative Filtering](https:\u002F\u002Fgithub.com\u002FcodeBehindMe\u002FCustomerIntelligence\u002Fblob\u002Fmaster\u002FCollaborativeFiltering.ipynb ) - Customer recommendation using collaborative filtering.\n* [Up-selling (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FExpected%20Value%20Business%20Model%20Performance\u002F ) - Analysis to identify up-selling opportunities. \n\n\n\u003Ca name=\"customer-cp\">\u003C\u002Fa>\n#### Churn Prediction\n\n* [Ride Sharing](https:\u002F\u002Fgithub.com\u002FMSopranoInTech\u002FChurn-prediction\u002Fblob\u002Fmaster\u002FChurn%20Prediction.ipynb) - Identify customer churn rates in order to target customers for retention campaigns. \n* [KKDBox I](https:\u002F\u002Fgithub.com\u002Fnaomifridman\u002FDeep-VAE-prediction-of-churn-customer) - Variational deep autoencoder to predict churn customer\n* [KKDBox II](https:\u002F\u002Fgithub.com\u002FFeaturetools\u002Fpredict-customer-churn) - A three step customer churn prediction framework using feature engineering. \n* [Personal Finance](https:\u002F\u002Fgithub.com\u002Fsmit5490\u002FCustomerChurn) - Predict customer subscription churn for a personal finance business. \n* [ANN](https:\u002F\u002Fgithub.com\u002FAgarwalGeeks\u002Fcustomer-churn-Analysis\u002Fblob\u002Fmaster\u002FANN.ipynb) - Churn analysis using artificial neural networks. \n* [Bike](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FCustomer%20Segmentation\u002F) - Customer bike churn analysis.\n* [Cost Sensitive](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Falbahnsen\u002FML_RiskManagement\u002Fblob\u002Fmaster\u002Fexercises\u002F10_CS_Churn.ipynb) - Cost sensitive churn analysis drivenby economic performance. \n\n\u003Ca name=\"customer-sent\">\u003C\u002Fa>\n#### Sentiment\n\n* [Topic Modelling](https:\u002F\u002Fgithub.com\u002FChrisjw42\u002FZLSurveyAnalysis) - Topic modelling on a corpus of customer surveys from the VR industry. \n* [Customer Satisfaction](https:\u002F\u002Fgithub.com\u002FBoulderDataScience\u002Fkaggle-santander) - Predict customer satisfaction using Kaggle data.\n\n\u003Ca name=\"employee\">\u003C\u002Fa>\n## Employee\n\n\u003Ca name=\"employee-man\">\u003C\u002Fa>\n#### Management\n* [Personality Prediction ](https:\u002F\u002Fgithub.com\u002Fjcl132\u002Fpersonality-prediction-from-text) - Predict Big 5 Personality from text. \n* [Salary Prediction Resume](https:\u002F\u002Fgithub.com\u002FArtifelse\u002FPrediction-salary-on-the-base-of-the-resume\u002Fblob\u002Fmaster\u002FNLP.ipynb) - Textual analyses over resume to predict appropriate salary [Project Disappeared, still a cool idea]\n* [Employee Review Analysis](https:\u002F\u002Fgithub.com\u002Fjackyip1\u002FIndeed-Reviews\u002Fblob\u002Fmaster\u002FPython%20scripts\u002FIndeed%20-%20Main.ipynb) - Review analytics for top 50 retail companies on Indeed.\n* [Diversity Analysis](https:\u002F\u002Fgithub.com\u002Fmtfaye\u002FEmployee-Diversity-in-Tech\u002Fblob\u002Fmaster\u002FData%20Viz%20Special%20Edition.ipynb) - A simple analysis of gender and race disparity in the tech industry.\n* [Occupation Prediction](https:\u002F\u002Fgithub.com\u002FRashmiSingh24\u002FOccuptionPrediction\u002Fblob\u002Fmaster\u002FBurningGlass.ipynb) - Predict the likelihood that an occupation is analytical. \n\n\n\u003Ca name=\"employee-perf\">\u003C\u002Fa>\n#### Performance\n* [Training Hours Performance](https:\u002F\u002Fgithub.com\u002Fniqueerdo\u002FMLpredictemployeedevelopment\u002Fblob\u002Fmaster\u002FWorking%20Ntbk_MODELS_Clustering.ipynb) - The impact of training ours on employee performance.\n* [Promotion Prediction](https:\u002F\u002Fgithub.com\u002FAbinSingh\u002FEmployee-Promotion-Prediction\u002Fblob\u002Fmaster\u002FEmployee_Promotion_Prediction.ipynb) - Analysing promotion patterns. \n* [Employee Attendance prediction](https:\u002F\u002Fgithub.com\u002Flokesh1233\u002FEmployee_Attendance\u002Ftree\u002Fmaster\u002Fnotebooks) - Various tools to predict employee attendance.\n\n\n\u003Ca name=\"employee-turn\">\u003C\u002Fa>\n#### Turnover\n* [Early Leaving Employees](https:\u002F\u002Fgithub.com\u002Fanushuk\u002FML-Human-Resources-Analytics\u002Fblob\u002Fmaster\u002FHuman%20Resources%20Analytics.ipynb ) - Identifying why the best and most experienced employees leaving prematurely.\n* [Employee Turnover](https:\u002F\u002Fgithub.com\u002Frandylaosat\u002FPredicting-Employee-Turnover-Complete-Guide-Analysis\u002Fblob\u002Fmaster\u002FHR%20Analytics%20Employee%20Turnover\u002FHR_Analytics_Report.ipynb) - Identifying factors associated with employee turnover.\n\n\n\u003Ca name=\"employee-con\">\u003C\u002Fa>\n#### Conversations\n* [Slack Communication Analysis](https:\u002F\u002Fgithub.com\u002Fstiebels\u002Fslack_nlp\u002Fblob\u002Fmaster\u002FSlack%20Analytics.ipynb) - Producing meaningful visualisations from slack conversations. \n* [Employee Relationships from Conversations ](https:\u002F\u002Fgithub.com\u002Fyuwie10\u002Fcultivate) - Identifying employee relationships from emails for improved HR analytics.\n* [Categorise Employee Requests](https:\u002F\u002Fgithub.com\u002Fdenizn\u002FRequest-classification-via-TFIDF) - Classifying employee requests via TFDIF Vectorizer and RandomForestClassifier.\n\n\n\u003Ca name=\"employee-ph\">\u003C\u002Fa>\n#### Physical\n* [Employee Face Recognition](https:\u002F\u002Fgithub.com\u002Fckarthic\u002FFace-Recognition) - A face recognition implementation. \n* [Attendance Management System](https:\u002F\u002Fgithub.com\u002Fmrsaicharan1\u002Fface-rec-a) - An attendance management system using face recognition.\n\n\u003Ca name=\"legal\">\u003C\u002Fa>\n## Legal\n\n\n\u003Ca name=\"legal-tools\">\u003C\u002Fa>\n#### Tools\n* [LexPredict](https:\u002F\u002Fgithub.com\u002FLexPredict\u002Flexpredict-contraxsuite ) - Software package and library. \n* [AI Para-legal](https:\u002F\u002Fgithub.com\u002Fdavidawad\u002Flobe) - Lobe is the world's first AI paralegal.\n* [Legal Entity Detection](https:\u002F\u002Fgithub.com\u002Fhockeyjudson\u002FLegal-Entity-Detection\u002Fblob\u002Fmaster\u002FDataset_conv.ipynb) - NER For Legal Documents.\n* [Legal Case Summarisation](https:\u002F\u002Fgithub.com\u002FLaw-AI\u002Fsummarization) - Implementation of different summarisation algorithms applied to legal case judgements.\n* [Legal Documents Google Scholar](https:\u002F\u002Fgithub.com\u002FGirrajMaheshwari\u002FWeb-scrapping-\u002Fblob\u002Fmaster\u002FGoogle_scholar%2BExtract%2Bcase%2Bdocument.ipynb ) - Using Google scholar to extract cases programatically. \n* [Chat Bot](https:\u002F\u002Fgithub.com\u002Fakarazeev\u002FLegalTech) - Chat-bot and email notifications.\n\n\n\u003Ca name=\"legal-pr\">\u003C\u002Fa>\n#### Policy and Regulatory\n* [GDPR scores](https:\u002F\u002Fgithub.com\u002Ferickjtorres\u002FAI_LegalDoc_Hackathon) - Predicting GDPR Scores for Legal Documents.\n* [Driving Factors FINRA](https:\u002F\u002Fgithub.com\u002Fsiddhantmaharana\u002Ftext-analysis-on-FINRA-docs) - Identify the driving factors that influence the FINRA arbitration decisions.\n* [Securities Bias Correction](https:\u002F\u002Fgithub.com\u002Fdavidsontheath\u002Fbias_corrected_estimators\u002Fblob\u002Fmaster\u002Fbias_corrected_estimators.ipynb ) - Bias-Corrected Estimation of Price Impact in Securities Litigation.\n* [Public Firm to Legal Decision](https:\u002F\u002Fgithub.com\u002Fanshu3769\u002FFirmEmbeddings) - Embed public firms based on their reaction to legal decisions.\n\n\n\u003Ca name=\"legal-judicial\">\u003C\u002Fa>\n#### Judicial Applied\n* [Supreme Court Prediction](https:\u002F\u002Fgithub.com\u002Fdavidmasse\u002FUS-supreme-court-prediction) - Predicting the ideological direction of Supreme Court decisions: ensemble vs. unified case-based model.\n* [Supreme Court Topic Modeling](https:\u002F\u002Fgithub.com\u002FAccelAI\u002FAI-Law-Minicourse\u002Ftree\u002Fmaster\u002FSupreme_Court_Topic_Modeling) - Multiple steps necessary to implement topic modeling on supreme court decisions. \n* [Judge Opinion](https:\u002F\u002Fgithub.com\u002FGirrajMaheshwari\u002FLegal-Analytics-project---Court-misclassification) - Using text mining and machine learning to analyze judges’ opinions for a particular concern. \n* [ML Law Matching](https:\u002F\u002Fgithub.com\u002Fwhs2k\u002FGPO-AI) - A machine learning law match maker.\n* [Bert Multi-label Classification](https:\u002F\u002Fgithub.com\u002Fbrightmart\u002Fsentiment_analysis_fine_grain) - Fine Grained Sentiment Analysis from AI.\n* [Some Computational AI Course](https:\u002F\u002Fwww.youtube.com\u002Fchannel\u002FUC5UHm2J9pbEZmWl97z_0hZw) - Video series Law MIT.\n\n\u003Ca name=\"management\">\u003C\u002Fa>\n## Management\n\n\u003Ca name=\"management-strat\">\u003C\u002Fa>\n#### Strategy\n* [Topic Model Reviews](https:\u002F\u002Fgithub.com\u002Fchrisjcc\u002FDataInsight\u002Fblob\u002Fmaster\u002FTopic_Analysis\u002FTopic_modeling_Amazon_Reviews.ipynb) - Amazon reviews for product development. \n* [Patents](https:\u002F\u002Fgithub.com\u002Fagdal1125\u002Fpatent_analysis) - Forecasting strategy using patents.\n* [Networks](https:\u002F\u002Fgithub.com\u002FJohnAnthonyBowllan\u002FBusinessAI\u002Fblob\u002Fmaster\u002FDataAnalysis_FeatureEngineering\u002FbusinessCommunitiesMethod.ipynb) - Business categories from Yelp reviews using networks can help to identify pockets of demand. \n* [Company Clustering](https:\u002F\u002Fgithub.com\u002FDistrictDataLabs\u002Fcompany-clustering) - Hierarchical clusters and topics from companies by extracting information from their descriptions on their websites\n* [Marketing Management](https:\u002F\u002Fgithub.com\u002FJiseong-Michael-Yang\u002FMarketing-Management) - Programmatic marketing management. \n\n\n\n\u003Ca name=\"management-do\">\u003C\u002Fa>\n#### Decision Optimisation\n* [Constraint Learning](https:\u002F\u002Fgithub.com\u002Fabrahami\u002FConstraint-Learning) - Machine learning that takes into account constraints. \n* [Fairlearn](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002Ffairlearn) - I think it is called cost-sensitive machine learning.\n* [Multi-label Classification](https:\u002F\u002Fgithub.com\u002Fej0cl6\u002Fcsmlc) - Cost-Sensitive Multi-Label Classification\n* [Multi-class Classification](https:\u002F\u002Fgithub.com\u002Fdavid-cortes\u002Fcostsensitive) - Cost-sensitive multi-class classification (Weighted-All-Pairs, Filter-Tree & others)\n* [CostCla](http:\u002F\u002Falbahnsen.github.io\u002FCostSensitiveClassification\u002F) - Costcla is a Python module for cost-sensitive machine learning (classification) built on top of Scikit-Learn\n* [DEA Software](https:\u002F\u002Faraith.github.io\u002FpyDEA\u002F) - pyDEA is a software package developed in Python for conducting data envelopment analysis (DEA).\n* [Covering Set (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FCovering%20Set\u002F) - Constraint programming analysis.\n* [Insurance (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FInsurance\u002F) - CP Insurance analysis.\n* [Machine Learning + CP (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FMachineLearningand%20Optimisation\u002F) - Machine Learning + Optimisation.\n* [Post Office (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FPost%20Office\u002F) - Post Office optimisation.\n* [Soda - CP (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fsoda_promotion-adapted-cp\u002F) - Constraint Programming + ML.\n* [Soda - Knapsack (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fsoda_promotion-adapted-knapsack\u002F) - Knapsack algorithm + ML.\n* [Soda - MLP (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fsoda_promotion-adapted-mip\u002F) - MLP analysis + ML.\n\n\u003Ca name=\"management-causal\">\u003C\u002Fa>\n#### Casual Inference\n* [Marketing AB Testing](https:\u002F\u002Fgithub.com\u002Fchrisjcc\u002FDataInsight\u002Ftree\u002Fmaster\u002FABtesting) - A\u002FB Testing Experiment.\n* [Legal Studies](https:\u002F\u002Fgithub.com\u002FAkesari12\u002FIntro_Causal_Inference) - Instrumental and discontinuity causal approach. \n* [A-B Test Result (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FAnalyze_ab_test_results\u002F) - Initial A-B Results.\n* [Causal Regression (FirmAI) ](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fcausal_regression\u002F) - Regression technique for causal estimate.\n* [Frequentist vs Bayesian A-B Test (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Ffrequentist-bayesian-ab-testing\u002F) - Comparison between frequentist and bayesian A-B testing.\n* [A-B Test Power Analysis (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FPower%20analysis%20for%20AB%20tests\u002F) - Sample size estimation to match testing power.\n* [Variance Reduction A-B test (FirmAI)](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fvariance-reduction\u002F) - Techniques to reduce variance in A-B tests.\n\n\n\u003Ca name=\"management-stat\">\u003C\u002Fa>\n#### Statistics\n* [Various](https:\u002F\u002Fgithub.com\u002Fkhanhnamle1994\u002Fapplied-machine-learning\u002Ftree\u002Fmaster\u002FStatistics) - Various applies statistical solutions \n\n\n\u003Ca name=\"management-quant\">\u003C\u002Fa>\n#### Quantitative \n* [Applied RL](https:\u002F\u002Fgithub.com\u002Fmimoralea\u002Fapplied-reinforcement-learning) - Reinforcement Learning and Decision Making tutorials explained at an intuitive level and with Jupyter Notebooks\n* [Process Mining](https:\u002F\u002Fgithub.com\u002Fyesanton\u002FProcess-Sequence-Prediction-with-A-priori-knowledge) - Leveraging A-priori Knowledge in Predictive Business Process Monitoring\n* [TS Forecasting](https:\u002F\u002Fgithub.com\u002Fkhanhnamle1994\u002Fapplied-machine-learning\u002Ftree\u002Fmaster\u002FTime-Series-Forecasting) - Time series forecasting for important business applications.\n\n####\n\n\u003Ca name=\"management-data\">\u003C\u002Fa>\n#### Data \n* [Web Scraping (FirmAI)](www.firmai.org\u002Fdata\u002F) - Web scraping solutions for Facebook, Glassdoor, Instagram, Morningstar, Similarweb, Yelp, Spyfu, Linkedin, Angellist. \n\n\n\u003Ca name=\"operations\">\u003C\u002Fa>\n## Operations\n\n\u003Ca name=\"operations-fail\">\u003C\u002Fa>\n#### Failure and Anomalies\n* [Anomalies](https:\u002F\u002Fgithub.com\u002Fyzhao062\u002Fanomaly-detection-resources) - Anomaly detection resources. \n* [Intrusion Detection](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Falbahnsen\u002FML_SecurityInformatics\u002Fblob\u002Fmaster\u002Fexercises\u002F05-IntrusionDetection.ipynb) - Detecting network intrusions. \n* [APS Failure](https:\u002F\u002Fgithub.com\u002FNisarg9795\u002FAnomaly-Detection-APS-failures-in-Scania-trucks\u002Fblob\u002Fmaster\u002F1_LR_Final_Code.py ), [Data](https:\u002F\u002Farchive.ics.uci.edu\u002Fml\u002Fdatasets\u002FAPS+Failure+at+Scania+Trucks) - Investigating APS failures in Scania trucks. \n* [Hardware Failure](https:\u002F\u002Fgithub.com\u002FAbertayMachineLearningGroup\u002Fmachine-learning-SIEM-water-infrastructure) - Using different machine learning techniques in detecting anomalies.\n* [Anomaly KIs](https:\u002F\u002Fgithub.com\u002Fhaowen-xu\u002Fdonut),[Paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.03903)  - Anomaly detection algorithm for seasonal KPIs.\n\n\u003Ca name=\"operations-load\">\u003C\u002Fa>\n#### Load and Capacity Management\n* [House Load Energy](https:\u002F\u002Fgithub.com\u002Fgiorgosfatouros\u002FAppliances-Energy-Load-Prediction) - Linear, SVR and Random Forest models to predict house's appliances energy Load.\n* [Uber Load Management](https:\u002F\u002Fgithub.com\u002Fbrianallen131\u002FUber-Predictive-Load-Management) - Uber predictive load management.\n* [Capacity Management](https:\u002F\u002Fgithub.com\u002Fnerdiejack\u002Fcapacity_management\u002Fblob\u002Fmaster\u002Fnotebooks\u002FMyWebshopAssignmentWithSolution.ipynb) - Investigating IT stability issues are caused by capacity constraints.\n* [Bike Sharing](https:\u002F\u002Fgithub.com\u002Fchrisjcc\u002FDataInsight\u002Fblob\u002Fmaster\u002FDataChallenge\u002FBikeShare_Challenge.ipynb) - XGBRegressor, RandomForestRegressor, GradientBoostingRegressor combined with feature selection.\n* [Airline Fleet Segmentation](http:\u002F\u002Fhtmlpreview.github.io\u002F?https:\u002F\u002Fgithub.com\u002Fatul-shukla-INSEAD\u002FGroupProjectBDA\u002Fblob\u002Fmaster\u002FGroupProject.html) - Analysis of Delta airlines.\n* [Airbnb](http:\u002F\u002Finseaddataanalytics.github.io\u002FINSEADAnalytics\u002Fgroupprojects\u002FAirbnbReport2016Jan.html) - Airbnb Booking Analysis.\n\n\n\u003Ca name=\"operations-predict\">\u003C\u002Fa>\n#### Prediction Management\n* [Dispute Prediction](https:\u002F\u002Fgithub.com\u002Fzhanghaizhen\u002FFinancial-Service-Complaint-Management\u002Ftree\u002Fmaster\u002Fipynb) - Financial service complaint management. \n* [Fight Delay Prediction](https:\u002F\u002Fgithub.com\u002Fcavaunpeu\u002Fflight-delays\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fflight-prediction.ipynb) - Transfer learning for flight-delay prediction via variational autoencoders in Keras.\n* [Electric Fault Prediction](https:\u002F\u002Fgithub.com\u002Fsusano0\u002FElectric-Fault-Prediction\u002Fblob\u002Fmaster\u002FFault_pred.ipynb) - Predict tripping at grid stations by applying simple machine learning algorithms.\n* [Popularity Prediction in R](https:\u002F\u002Fgithub.com\u002Fs-mishra\u002Ffeaturedriven-hawkes\u002Fblob\u002Fmaster\u002Fcode\u002Fmarked_hawkes_point_process.ipynb) - Marked Hawkes Point Process .\n","# 商业机器学习与数据科学应用\n\n---------\n\n\n## 🌟 我们正在壮大！\n\n我们计划在2024年与积极进取、独立自主的博士毕业生或在读博士生合作开展约七项新项目。如果您有兴趣参与前沿的投资洞察和数据分析工作，请随时联系我们！合作形式可以是与高校联合进行，也可以作为独立研究。\n\n![image](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ffirmai_business-machine-learning_readme_dbe070c495d5.png)\n\n\n### 🚀 关于Sov.ai\n\nSov.ai致力于将先进的机器学习技术与金融数据分析相结合，以革新投资策略。我们目前正与**排名前十的量化对冲基金中的三家**以及众多中型和精品金融机构合作。\n\n我们的平台利用多元化的数据源和创新算法，提供可操作的洞察，助力更明智的投资决策。加入Sov.ai，您将成为一支充满活力的研究团队的一员，共同通过技术突破金融领域的边界。在表达您的兴趣之前，请知悉这些研究将以高挑战性和实验性为主。\n\n\n### 🔍 研究与项目机会\n\n我们提供多种项目，涵盖机器学习和金融领域的不同兴趣与专长。近期一些令人兴奋的项目包括：\n\n- **基于GitHub日志的预测建模：** 利用GitHub活动和开发者数据构建模型，以预测市场趋势和投资机会。\n- **卫星数据分析：** 探索社交媒体情绪、卫星图像或网络流量等非传统数据源，以提升财务预测能力。\n- **数据插补技术：** 研究处理缺失或不完整数据的新方法，从而提高模型的稳健性和准确性。\n\n更多已集成到订阅产品中的公开项目信息，请访问[docs.sov.ai](https:\u002F\u002Fdocs.sov.ai)。如果您已有企业赞助方，我们也非常乐意与其合作。\n\n\n### 🌐 为什么加入Sov.ai？\n\n- **创新环境：** 接触机器学习和金融领域最新的技术和方法论。\n- **协作团队：** 与一群热衷于推动投资洞察创新的专家共事。\n- **灵活项目：** 根据您的兴趣和专长定制研究方向，并自由探索新想法。\n- **资深研究人员：** 团队成员曾就职于纽约大学、哥伦比亚大学、牛津曼彻斯特学院、艾伦·图灵研究所及剑桥大学等知名机构。\n- **职业发展：** 与已加入DRW、Citadel Securities、Virtu Financial、Akuna Capital、HRT等公司的校友建立联系。\n\n\n### 🤝 如何申请\n\n如果您渴望运用自己在机器学习和金融领域的专业知识，开展具有影响力的科研与项目，我们非常期待与您交流！请将您的简历及简要的研究兴趣说明发送至 [research@sov.ai](mailto:research@sov.ai)。\n\n加入我们，共同塑造投资洞察的未来，在金融领域产生深远影响！\n\n\n\n## 目录\n\n### 部门应用\n\u003C!-- MarkdownTOC depth=4 -->\n\n- [会计](#accounting)\n    - [机器学习](#accounting-ml)\n    - [分析](#accounting-analytics)\n    - [文本分析](#accounting-text)\n    - [数据](#accounting-data)\n    - [研究与文章](#accounting-ra)\n    - [网站](#accounting-web)\n    - [课程](#accounting-course)\n- [客户](#customer)\n    - [客户终身价值](#customer-clv)\n    - [细分](#customer-seg)\n    - [行为](#customer-behave)\n    - [推荐系统](#customer-rec)\n    - [流失预测](#customer-cp)\n    - [情感分析](#customer-sent)\n- [员工](#employee)\n    - [管理](#employee-man)\n    - [绩效](#employee-perf)\n    - [人员流动](#employee-general-turn)\n    - [对话分析](#employee-con)\n    - [物理相关](#employee-ph)\n- [法律](#legal)\n    - [工具](#legal-tools)\n    - [政策与监管](#legal-pr)\n    - [司法](#legal-judicial)\n- [管理](#management)\n    - [战略](#management-strat)\n    - [决策优化](#management-do)\n    - [因果推断](#management-causal)\n    - [统计](#management-stat)\n    - [量化](#management-quant)\n    - [数据](#management-data)\n- [运营](#operations)\n    - [故障与异常](#operations-fail)\n    - [负载与容量管理](#operations-load)\n    - [预测管理](#operations-predict)\n\n\n\u003C!-- \u002FMarkdownTOC -->\n#### 更多内容请参阅 [Python商业分析](https:\u002F\u002Fgithub.com\u002Ffirmai\u002Fpython-business-analytics)\n\n\u003Ca name=\"accounting\">\u003C\u002Fa>\n\n## 会计\n\n\u003Ca name=\"accounting-ml\">\u003C\u002Fa>\n#### 机器学习\n* [会计科目预测](https:\u002F\u002Fgithub.com\u002Fagdgovsg\u002Fml-coa-charging ) - 使用标注数据为每笔交易建议合适的科目名称。\n* [会计异常检测](https:\u002F\u002Fgithub.com\u002FGitiHubi\u002FdeepAI\u002Fblob\u002Fmaster\u002FGTC_2018_Lab-solutions.ipynb) - 利用深度学习框架识别会计异常。\n* [财务报表异常检测](https:\u002F\u002Fgithub.com\u002Frameshcalamur\u002Ffin-stmt-anom) - 使用R语言在提交前检测异常。\n* [资产使用寿命预测（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FAged%20Debtors\u002F) - 基于传感器观测和特征工程预测资产的使用寿命。\n* [人工智能应用于XBRL](https:\u002F\u002Fgithub.com\u002FNiels-Peter\u002FXBRL-AI) - 将XBRL标准化表示转换为适用于人工智能和机器学习的形式。\n\n\u003Ca name=\"accounting-analytics\">\u003C\u002Fa>\n#### 分析\n* [法务会计](https:\u002F\u002Fgithub.com\u002Fmschermann\u002Fforensic_accounting) - 收集利用数据分析进行法务会计的案例研究。正在寻找更多数据以实践法务会计，请[联系](https:\u002F\u002Fgithub.com\u002Fmschermann\u002F)。\n* [总账（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FGeneral%20Ledger\u002F) - 对通过会计系统导出的总账数据进行处理。\n* [子弹图（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FBullet-Graph-Article\u002F) - 子弹图可视化有助于跟踪销售、佣金及其他绩效指标。\n* [逾期应收账款分析（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FAged%20Debtors\u002F) - 示例分析，用于调查逾期应收账款。\n* [自动化财务报表XBRL](https:\u002F\u002Fgithub.com\u002FCharlesHoffmanCPA\u002FcharleshoffmanCPA.github.io) - 使用XML语言，并可将分析结果导入Python。\n\n\u003Ca name=\"accounting-text\">\u003C\u002Fa>\n#### 文本分析\n* [财务情绪分析](https:\u002F\u002Fgithub.com\u002FEricHe98\u002FFinancial-Statements-Text-Analysis) - 通过情感、距离和比例分析生成交易信号。\n* [全面自然语言处理](https:\u002F\u002Fgithub.com\u002FTiesdeKok\u002FPython_NLP_Tutorial\u002Fblob\u002Fmaster\u002FNLP_Notebook.ipynb) - 适用于会计研究的综合NLP技术。\n\n\u003Ca name=\"accounting-data\">\u003C\u002Fa>\n#### 数据、解析与API\n* [EDGAR](https:\u002F\u002Fgithub.com\u002FTiesdeKok\u002FUW_Python_Camp\u002Fblob\u002Fmaster\u002FMaterials\u002FSession_5\u002FEDGAR_walkthrough.ipynb) - EDGAR数据获取指南。\n* [IRS](http:\u002F\u002Fsocial-metrics.org\u002Fsox\u002F) - 访问和解析IRS申报文件。\n* [企业财务数据](http:\u002F\u002Fraw.rutgers.edu\u002FCorporate%20Financial%20Data.html) - 罗格斯大学的企业财务数据集。\n* [非财务企业数据](http:\u002F\u002Fraw.rutgers.edu\u002FNon-Financial%20Corporate%20Data.html) - 罗格斯大学的非财务企业数据集。\n* [PDF解析](https:\u002F\u002Fgithub.com\u002Fdanshorstein\u002Fpython4cpas\u002Fblob\u002Fmaster\u002F03_parsing_pdf_files\u002FAR%20Aging%20-%20working.ipynb) - 从PDF文档中提取有用数据。\n* [PDF表格转Excel](https:\u002F\u002Fgithub.com\u002Fdanshorstein\u002Fficpa_article) - 如何将PDF中的表格输出为Excel文件。\n\n\u003Ca name=\"accounting-ra\">\u003C\u002Fa>\n#### 研究与文章\n* [理解会计分析](http:\u002F\u002Fsocial-metrics.org\u002Faccountinganalytics\u002F) - 一篇文章探讨了会计分析的重要性。\n* [VLFeat](http:\u002F\u002Fwww.vlfeat.org\u002F) - VLFeat是一个开源且跨平台的计算机视觉算法库，提供Matlab工具箱。\n\n\u003Ca name=\"accounting-web\">\u003C\u002Fa>\n#### 网站\n* [罗格斯Raw](http:\u002F\u002Fraw.rutgers.edu\u002F) - 来自罗格斯大学的优质数字会计研究资源。\n\n\u003Ca name=\"accounting-course\">\u003C\u002Fa>\n#### 课程\n* [计算机辅助会计](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLauepKFT6DK8TaNaq_SqZW4LIDJhCkZe2) - 罗格斯大学的一系列视频，探讨如何利用计算技术改进会计工作。\n* [数字化时代的会计](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLauepKFT6DK8_Xun584UQPPsg1grYkWw0) - 罗格斯大学的另一系列视频，研究数字化时代对会计的影响。\n\n\u003Ca name=\"customer\">\u003C\u002Fa>\n\n## 客户\n\n\u003Ca name=\"customer-clv\">\u003C\u002Fa>\n#### 生命周期价值\n* [帕累托\u002FNBD模型](https:\u002F\u002Fgithub.com\u002Fzabahana\u002FCustomer-LifeTime-Value-Analysis\u002Fblob\u002Fmaster\u002FCLV%20Analysis.ipynb) - 使用帕累托\u002FNBD模型计算客户生命周期价值。\n* [队列分析](https:\u002F\u002Fgithub.com\u002Firis9112\u002FCustomer-Segmentation\u002Fblob\u002Fmaster\u002FChapter1-Cohort_Analysis.ipynb) - 队列分析，按时间维度将客户划分为互斥的队列。\n\n \n\u003Ca name=\"customer-seg\">\u003C\u002Fa>\n#### 客户细分\n\n* [电子商务](https:\u002F\u002Fgithub.com\u002Fjalajthanaki\u002FCustomer_segmentation\u002Fblob\u002Fmaster\u002FCust_segmentation_online_retail.ipynb ) - 电子商务客户细分。\n* [杂货店](https:\u002F\u002Fgithub.com\u002Fharry329\u002FCustomerFinding\u002Fblob\u002Fmaster\u002Fcustomer_segments.ipynb ) - 杂货店客户细分。\n* [在线零售商](https:\u002F\u002Fgithub.com\u002FVinayak02\u002FCustomerCentricRetail\u002Fblob\u002Fmaster\u002FCustomerSegmentation\u002FCustomer_Segmentation_Online_Retail.ipynb) - 在线零售商客户细分。\n* [银行](https:\u002F\u002Fgithub.com\u002FMogbo\u002FCustomer-Clustering-Segmentation-Challenge) - 银行客户细分。\n* [批发客户](https:\u002F\u002Fgithub.com\u002FSyedAdilAli93\u002FIdentifying-Customers\u002Fblob\u002Fmaster\u002Fcustomer_segments.ipynb) - 批发客户的聚类分析。\n* [多种方法](https:\u002F\u002Fgithub.com\u002Fabalaji-blr\u002FCustomerSegments\u002Ftree\u002Fmaster\u002Fdeliver ) - 多种类型的细分和聚类技术。\n\n\n\u003Ca name=\"customer-behave\">\u003C\u002Fa>\n#### 行为分析\n\n* [RNN](https:\u002F\u002Fgithub.com\u002FDaniSanchezSantolaya\u002FRNN-customer-behavior\u002Ftree\u002Fmaster\u002Fsrc) - 使用RNN模型进行序列分析，研究客户随时间的行为变化。\n* [神经网络](https:\u002F\u002Fgithub.com\u002FVinayak02\u002FCustomerCentricRetail\u002Fblob\u002Fmaster\u002FDemandForecasting\u002FNeuralNetworks.ipynb) - 使用人工神经网络进行需求预测。\n* [时间序列分析](https:\u002F\u002Fgithub.com\u002Friccotti\u002FCustomerTemporalRegularities) - 研究客户的周期性行为模式。\n* [POS数据分析](https:\u002F\u002Fgithub.com\u002FIBM\u002Fcustomer_pos_analytics\u002Fblob\u002Fmaster\u002Fcode\u002FCustomer%20Ranking%20POS%20wip.ipynb) - 基于POS数据，通过分析驱动零售促销中的客户行为排名。\n* [批发客户](https:\u002F\u002Fgithub.com\u002Fkralmachine\u002FWholesaleCustomerAnalysis\u002Fblob\u002Fmaster\u002FWhosaleCustomerAnalysis.ipynb) - 批发客户的探索性数据分析。\n* [RFM](https:\u002F\u002Fgithub.com\u002Fespin086\u002Fcustomer_growth\u002Fblob\u002Fmaster\u002Frfm\u002Frfm.ipynb) - 进行RFM（最近购买、购买频率、消费金额）分析。\n* [退货行为](https:\u002F\u002Fgithub.com\u002Fadarsh2111\u002FCustomer-Returns-Analysis-Customer-Fraud-Detection-\u002Fblob\u002Fmaster\u002FReturns%20Analysis.ipynb) - 预测总退货量及欺诈性退货。\n* [访问次数](https:\u002F\u002Fgithub.com\u002FRyanfras\u002FCustomer-Visits\u002Fblob\u002Fmaster\u002FCustomer%20Visits.ipynb) - 预测客户每周哪一天会来访。\n* [银行：下一次购买](https:\u002F\u002Fgithub.com\u002Falbertcdc\u002FProject_CAJAMAR) - 一个预测银行客户最可能下一次购买的项目。\n* [银行：目标客户预测](https:\u002F\u002Fgithub.com\u002Frohangawade\u002FPredicting-Target-customers-for-Bank-Policy-subscribtion-using-Logistic_Regression_Transparency) - 预测将订阅银行新政策的目标客户。\n* [下一次购买](https:\u002F\u002Fgithub.com\u002FFeaturetools\u002Fpredict-next-purchase) - 同样使用特征工程来预测客户的下一次购买。\n* [客户重复购买](https:\u002F\u002Fgithub.com\u002Fkpei\u002FCustomer-Analytics\u002Fblob\u002Fmaster\u002Fcustomer_zakka.ipynb) - 使用lifetimes Python库和真实的珠宝零售商数据，分析客户的重复购买行为。\n* [A\u002FB测试](https:\u002F\u002Fgithub.com\u002Fsushant2811\u002FcustomerAnalyticsWithA-BTesting\u002Fblob\u002Fmaster\u002FcustomerAnalyticsWithA-BTesting.ipynb) - 找到最佳KPI并进行A\u002FB测试。\n* [客户调查（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FCustomer%20Survey\u002F) - 解析和分析客户调查的示例。\n* [幸福感](https:\u002F\u002Fgithub.com\u002Frohit6205\u002FpredictHappiness\u002Fblob\u002Fmaster\u002FpredictingHapiness.ipynb) - 利用酒店顾客评价分析其入住期间的幸福感。\n* [其他客户分析](https:\u002F\u002Fgithub.com\u002Fmapr-demos\u002Fcustomer360) - 各种用于客户分析的工具和技术。\n\n\n\u003Ca name=\"customer-rec\">\u003C\u002Fa>\n#### 推荐系统\n\n* [推荐](https:\u002F\u002Fgithub.com\u002Fannalucia1\u002FCustomer-Behavior-Analysis-Recommendation\u002Fblob\u002Fmaster\u002Frecomendation_by_RatingScore.ipynb) - 根据音乐应用用户的评分推荐他们可能喜欢的歌曲。\n* [通用推荐系统](https:\u002F\u002Fgithub.com\u002FVinayak02\u002FCustomerCentricRetail\u002Fblob\u002Fmaster\u002FRecommenderSystem\u002FRecommender.ipynb) - 确定向哪些客户推荐哪些产品。\n* [协同过滤](https:\u002F\u002Fgithub.com\u002FcodeBehindMe\u002FCustomerIntelligence\u002Fblob\u002Fmaster\u002FCollaborativeFiltering.ipynb ) - 使用协同过滤进行客户推荐。\n* [追加销售（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FExpected%20Value%20Business%20Model%20Performance\u002F ) - 分析以识别追加销售机会。\n\n\n\u003Ca name=\"customer-cp\">\u003C\u002Fa>\n#### 客户流失预测\n\n* [网约车](https:\u002F\u002Fgithub.com\u002FMSopranoInTech\u002FChurn-prediction\u002Fblob\u002Fmaster\u002FChurn%20Prediction.ipynb) - 识别客户流失率，以便针对客户开展留存活动。\n* [KKDBox I](https:\u002F\u002Fgithub.com\u002Fnaomifridman\u002FDeep-VAE-prediction-of-churn-customer) - 使用变分深度自编码器预测流失客户。\n* [KKDBox II](https:\u002F\u002Fgithub.com\u002FFeaturetools\u002Fpredict-customer-churn) - 一个基于特征工程的三步客户流失预测框架。\n* [个人理财](https:\u002F\u002Fgithub.com\u002Fsmit5490\u002FCustomerChurn) - 预测个人理财业务中的客户订阅流失情况。\n* [ANN](https:\u002F\u002Fgithub.com\u002FAgarwalGeeks\u002Fcustomer-churn-Analysis\u002Fblob\u002Fmaster\u002FANN.ipynb) - 使用人工神经网络进行流失分析。\n* [自行车](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FCustomer%20Segmentation\u002F) - 客户自行车流失分析。\n* [成本敏感型](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Falbahnsen\u002FML_RiskManagement\u002Fblob\u002Fmaster\u002Fexercises\u002F10_CS_Churn.ipynb) - 基于经济绩效的成本敏感型流失分析。\n\n\u003Ca name=\"customer-sent\">\u003C\u002Fa>\n#### 情感分析\n\n* [主题建模](https:\u002F\u002Fgithub.com\u002FChrisjw42\u002FZLSurveyAnalysis) - 对VR行业客户调查语料库进行主题建模。\n* [客户满意度](https:\u002F\u002Fgithub.com\u002FBoulderDataScience\u002Fkaggle-santander) - 使用Kaggle数据预测客户满意度。\n\n\u003Ca name=\"employee\">\u003C\u002Fa>\n\n## 员工\n\n\u003Ca name=\"employee-man\">\u003C\u002Fa>\n#### 管理\n* [人格预测](https:\u002F\u002Fgithub.com\u002Fjcl132\u002Fpersonality-prediction-from-text) - 从文本中预测大五人格。\n* [基于简历的薪资预测](https:\u002F\u002Fgithub.com\u002FArtifelse\u002FPrediction-salary-on-the-base-of-the-resume\u002Fblob\u002Fmaster\u002FNLP.ipynb) - 对简历进行文本分析以预测合适的薪资。[项目已消失，但想法依然很酷]\n* [员工评论分析](https:\u002F\u002Fgithub.com\u002Fjackyip1\u002FIndeed-Reviews\u002Fblob\u002Fmaster\u002FPython%20scripts\u002FIndeed%20-%20Main.ipynb) - 对Indeed上排名前50的零售公司进行评论分析。\n* [多样性分析](https:\u002F\u002Fgithub.com\u002Fmtfaye\u002FEmployee-Diversity-in-Tech\u002Fblob\u002Fmaster\u002FData%20Viz%20Special%20Edition.ipynb) - 对科技行业中性别和种族差异的简单分析。\n* [职业预测](https:\u002F\u002Fgithub.com\u002FRashmiSingh24\u002FOccuptionPrediction\u002Fblob\u002Fmaster\u002FBurningGlass.ipynb) - 预测某一职业是否具有分析性。\n\n\n\u003Ca name=\"employee-perf\">\u003C\u002Fa>\n#### 绩效\n* [培训时长对绩效的影响](https:\u002F\u002Fgithub.com\u002Fniqueerdo\u002FMLpredictemployeedevelopment\u002Fblob\u002Fmaster\u002FWorking%20Ntbk_MODELS_Clustering.ipynb) - 分析培训时长对员工绩效的影响。\n* [晋升预测](https:\u002F\u002Fgithub.com\u002FAbinSingh\u002FEmployee-Promotion-Prediction\u002Fblob\u002Fmaster\u002FEmployee_Promotion_Prediction.ipynb) - 分析员工晋升模式。\n* [员工出勤率预测](https:\u002F\u002Fgithub.com\u002Flokesh1233\u002FEmployee_Attendance\u002Ftree\u002Fmaster\u002Fnotebooks) - 使用多种工具预测员工出勤情况。\n\n\n\u003Ca name=\"employee-turn\">\u003C\u002Fa>\n#### 流动率\n* [提前离职员工分析](https:\u002F\u002Fgithub.com\u002Fanushuk\u002FML-Human-Resources-Analytics\u002Fblob\u002Fmaster\u002FHuman%20Resources%20Analytics.ipynb) - 识别优秀且经验丰富的员工为何会过早离职。\n* [员工流动率预测](https:\u002F\u002Fgithub.com\u002Frandylaosat\u002FPredicting-Employee-Turnover-Complete-Guide-Analysis\u002Fblob\u002Fmaster\u002FHR%20Analytics%20Employee%20Turnover\u002FHR_Analytics_Report.ipynb) - 识别与员工流动率相关的因素。\n\n\n\u003Ca name=\"employee-con\">\u003C\u002Fa>\n#### 沟通\n* [Slack沟通分析](https:\u002F\u002Fgithub.com\u002Fstiebels\u002Fslack_nlp\u002Fblob\u002Fmaster\u002FSlack%20Analytics.ipynb) - 从Slack对话中生成有意义的可视化图表。\n* [通过邮件识别员工关系](https:\u002F\u002Fgithub.com\u002Fyuwie10\u002Fcultivate) - 从电子邮件中识别员工关系，以改进人力资源分析。\n* [员工请求分类](https:\u002F\u002Fgithub.com\u002Fdenizn\u002FRequest-classification-via-TFIDF) - 使用TF-IDF向量化器和随机森林分类器对员工请求进行分类。\n\n\n\u003Ca name=\"employee-ph\">\u003C\u002Fa>\n#### 物理\n* [员工人脸识别](https:\u002F\u002Fgithub.com\u002Fckarthic\u002FFace-Recognition) - 人脸识别实现。\n* [考勤管理系统](https:\u002F\u002Fgithub.com\u002Fmrsaicharan1\u002Fface-rec-a) - 基于人脸识别的考勤管理系统。\n\n\u003Ca name=\"legal\">\u003C\u002Fa>\n## 法律\n\n\n\u003Ca name=\"legal-tools\">\u003C\u002Fa>\n#### 工具\n* [LexPredict](https:\u002F\u002Fgithub.com\u002FLexPredict\u002Flexpredict-contraxsuite) - 软件包和库。\n* [AI律师助理](https:\u002F\u002Fgithub.com\u002Fdavidawad\u002Flobe) - Lobe是全球首个AI律师助理。\n* [法律实体识别](https:\u002F\u002Fgithub.com\u002Fhockeyjudson\u002FLegal-Entity-Detection\u002Fblob\u002Fmaster\u002FDataset_conv.ipynb) - 面向法律文件的命名实体识别。\n* [法律案件摘要](https:\u002F\u002Fgithub.com\u002FLaw-AI\u002Fsummarization) - 将不同摘要算法应用于法律案件判决的实现。\n* [法律文献Google Scholar爬取](https:\u002F\u002Fgithub.com\u002FGirrajMaheshwari\u002FWeb-scrapping-\u002Fblob\u002Fmaster\u002FGoogle_scholar%2BExtract%2Bcase%2Bdocument.ipynb) - 利用Google Scholar程序化提取案例文档。\n* [聊天机器人](https:\u002F\u002Fgithub.com\u002Fakarazeev\u002FLegalTech) - 聊天机器人及邮件通知。\n\n\n\u003Ca name=\"legal-pr\">\u003C\u002Fa>\n#### 政策与监管\n* [GDPR评分](https:\u002F\u002Fgithub.com\u002Ferickjtorres\u002FAI_LegalDoc_Hackathon) - 预测法律文件的GDPR合规评分。\n* [影响FINRA裁决的因素](https:\u002F\u002Fgithub.com\u002Fsiddhantmaharana\u002Ftext-analysis-on-FINRA-docs) - 识别影响FINRA仲裁决定的关键因素。\n* [证券诉讼中的偏差校正估计](https:\u002F\u002Fgithub.com\u002Fdavidsontheath\u002Fbias_corrected_estimators\u002Fblob\u002Fmaster\u002Fbias_corrected_estimators.ipynb) - 在证券诉讼中对价格影响进行偏差校正估计。\n* [上市公司对法律判决的反应嵌入](https:\u002F\u002Fgithub.com\u002Fanshu3769\u002FFirmEmbeddings) - 根据上市公司对法律判决的反应对其进行嵌入表示。\n\n\n\u003Ca name=\"legal-judicial\">\u003C\u002Fa>\n#### 司法应用\n* [美国最高法院判决方向预测](https:\u002F\u002Fgithub.com\u002Fdavidmasse\u002FUS-supreme-court-prediction) - 使用集成模型与统一的基于案件的模型预测最高法院判决的意识形态倾向。\n* [最高法院主题建模](https:\u002F\u002Fgithub.com\u002FAccelAI\u002FAI-Law-Minicourse\u002Ftree\u002Fmaster\u002FSupreme_Court_Topic_Modeling) - 实现最高法院判决主题建模所需的多个步骤。\n* [法官意见分析](https:\u002F\u002Fgithub.com\u002FGirrajMaheshwari\u002FLegal-Analytics-project---Court-misclassification) - 利用文本挖掘和机器学习分析法官针对特定问题的意见。\n* [ML法律匹配](https:\u002F\u002Fgithub.com\u002Fwhs2k\u002FGPO-AI) - 一款基于机器学习的法律匹配工具。\n* [BERT多标签分类](https:\u002F\u002Fgithub.com\u002Fbrightmart\u002Fsentiment_analysis_fine_grain) - 基于AI的细粒度情感分析。\n* [一些计算AI课程](https:\u002F\u002Fwww.youtube.com\u002Fchannel\u002FUC5UHm2J9pbEZmWl97z_0hZw) - MIT法学视频系列。\n\n\u003Ca name=\"management\">\u003C\u002Fa>\n\n## 管理\n\n\u003Ca name=\"management-strat\">\u003C\u002Fa>\n#### 战略\n* [主题模型评论](https:\u002F\u002Fgithub.com\u002Fchrisjcc\u002FDataInsight\u002Fblob\u002Fmaster\u002FTopic_Analysis\u002FTopic_modeling_Amazon_Reviews.ipynb) - 用于产品开发的亚马逊评论分析。\n* [专利](https:\u002F\u002Fgithub.com\u002Fagdal1125\u002Fpatent_analysis) - 利用专利进行预测策略。\n* [网络](https:\u002F\u002Fgithub.com\u002FJohnAnthonyBowllan\u002FBusinessAI\u002Fblob\u002Fmaster\u002FDataAnalysis_FeatureEngineering\u002FbusinessCommunitiesMethod.ipynb) - 使用Yelp评论中的网络数据识别商业类别，有助于发现潜在需求。\n* [公司聚类](https:\u002F\u002Fgithub.com\u002FDistrictDataLabs\u002Fcompany-clustering) - 通过提取公司官网描述信息，构建层次聚类和主题模型。\n* [营销管理](https:\u002F\u002Fgithub.com\u002FJiseong-Michael-Yang\u002FMarketing-Management) - 程序化营销管理。\n\n\n\n\u003Ca name=\"management-do\">\u003C\u002Fa>\n#### 决策优化\n* [约束学习](https:\u002F\u002Fgithub.com\u002Fabrahami\u002FConstraint-Learning) - 考虑约束条件的机器学习方法。\n* [Fairlearn](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002Ffairlearn) - 这是一种成本敏感型机器学习方法。\n* [多标签分类](https:\u002F\u002Fgithub.com\u002Fej0cl6\u002Fcsmlc) - 成本敏感型多标签分类。\n* [多分类问题](https:\u002F\u002Fgithub.com\u002Fdavid-cortes\u002Fcostsensitive) - 成本敏感型多分类问题（加权全对、滤波树等）。\n* [Costcla](http:\u002F\u002Falbahnsen.github.io\u002FCostSensitiveClassification\u002F) - Costcla是一个基于Scikit-Learn的Python模块，用于成本敏感型机器学习（分类）。\n* [DEA软件](https:\u002F\u002Faraith.github.io\u002FpyDEA\u002F) - pyDEA是用Python开发的数据包络分析（DEA）软件。\n* [覆盖集（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FCovering%20Set\u002F) - 约束编程分析。\n* [保险（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FInsurance\u002F) - CP保险分析。\n* [机器学习+CP（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FMachineLearningand%20Optimisation\u002F) - 机器学习与优化结合。\n* [邮局（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FPost%20Office\u002F) - 邮局优化问题。\n* [汽水 - CP（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fsoda_promotion-adapted-cp\u002F) - 约束编程 + 机器学习。\n* [汽水 - 背包问题（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fsoda_promotion-adapted-knapsack\u002F) - 背包算法 + 机器学习。\n* [汽水 - MLP（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fsoda_promotion-adapted-mip\u002F) - MLP分析 + 机器学习。\n\n\u003Ca name=\"management-causal\">\u003C\u002Fa>\n#### 因果推断\n* [营销AB测试](https:\u002F\u002Fgithub.com\u002Fchrisjcc\u002FDataInsight\u002Ftree\u002Fmaster\u002FABtesting) - A\u002FB测试实验。\n* [法律研究](https:\u002F\u002Fgithub.com\u002FAkesari12\u002FIntro_Causal_Inference) - 工具变量法和断点回归法的因果推断方法。\n* [A-B测试结果（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FAnalyze_ab_test_results\u002F) - 初步的A-B测试结果。\n* [因果回归（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fcausal_regression\u002F) - 用于因果估计的回归技术。\n* [频率派与贝叶斯派A-B测试（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Ffrequentist-bayesian-ab-testing\u002F) - 频率派与贝叶斯派A-B测试的比较。\n* [A-B测试功效分析（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002FPower%20analysis%20for%20AB%20tests\u002F) - 估算样本量以匹配测试功效。\n* [方差缩减A-B测试（FirmAI）](http:\u002F\u002Fwww.firmai.org\u002Fdocuments\u002Fvariance-reduction\u002F) - 降低A-B测试中方差的技术。\n\n\n\u003Ca name=\"management-stat\">\u003C\u002Fa>\n#### 统计学\n* [各种应用](https:\u002F\u002Fgithub.com\u002Fkhanhnamle1994\u002Fapplied-machine-learning\u002Ftree\u002Fmaster\u002FStatistics) - 各种统计学解决方案的应用。\n\n\n\u003Ca name=\"management-quant\">\u003C\u002Fa>\n#### 定量分析\n* [应用强化学习](https:\u002F\u002Fgithub.com\u002Fmimoralea\u002Fapplied-reinforcement-learning) - 以直观方式讲解并配有Jupyter Notebook的强化学习与决策教程。\n* [流程挖掘](https:\u002F\u002Fgithub.com\u002Fyesanton\u002FProcess-Sequence-Prediction-with-A-priori-knowledge) - 在预测性业务流程监控中利用先验知识。\n* [时间序列预测](https:\u002F\u002Fgithub.com\u002Fkhanhnamle1994\u002Fapplied-machine-learning\u002Ftree\u002Fmaster\u002FTime-Series-Forecasting) - 针对重要业务场景的时间序列预测。\n\n####\n\n\u003Ca name=\"management-data\">\u003C\u002Fa>\n#### 数据\n* [网页爬取（FirmAI）](www.firmai.org\u002Fdata\u002F) - 针对Facebook、Glassdoor、Instagram、Morningstar、Similarweb、Yelp、Spyfu、LinkedIn、AngelList等平台的数据抓取解决方案。\n\n\n\u003Ca name=\"operations\">\u003C\u002Fa>\n\n## 运营\n\n\u003Ca name=\"operations-fail\">\u003C\u002Fa>\n#### 故障与异常\n* [异常检测](https:\u002F\u002Fgithub.com\u002Fyzhao062\u002Fanomaly-detection-resources) - 异常检测资源。\n* [入侵检测](https:\u002F\u002Fnbviewer.jupyter.org\u002Fgithub\u002Falbahnsen\u002FML_SecurityInformatics\u002Fblob\u002Fmaster\u002Fexercises\u002F05-IntrusionDetection.ipynb) - 检测网络入侵。\n* [APS故障](https:\u002F\u002Fgithub.com\u002FNisarg9795\u002FAnomaly-Detection-APS-failures-in-Scania-trucks\u002Fblob\u002Fmaster\u002F1_LR_Final_Code.py)，[数据](https:\u002F\u002Farchive.ics.uci.edu\u002Fml\u002Fdatasets\u002FAPS+Failure+at+Scania+Trucks) - 研究斯堪尼亚卡车中的APS故障。\n* [硬件故障](https:\u002F\u002Fgithub.com\u002FAbertayMachineLearningGroup\u002Fmachine-learning-SIEM-water-infrastructure) - 使用不同的机器学习技术来检测异常。\n* [异常KPI](https:\u002F\u002Fgithub.com\u002Fhaowen-xu\u002Fdonut)，[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.03903) - 用于季节性KPI的异常检测算法。\n\n\u003Ca name=\"operations-load\">\u003C\u002Fa>\n#### 负载与容量管理\n* [家庭用电负荷](https:\u002F\u002Fgithub.com\u002Fgiorgosfatouros\u002FAppliances-Energy-Load-Prediction) - 使用线性模型、SVR和随机森林模型预测家庭电器的用电负荷。\n* [优步负载管理](https:\u002F\u002Fgithub.com\u002Fbrianallen131\u002FUber-Predictive-Load-Management) - 优步预测性负载管理。\n* [容量管理](https:\u002F\u002Fgithub.com\u002Fnerdiejack\u002Fcapacity_management\u002Fblob\u002Fmaster\u002Fnotebooks\u002FMyWebshopAssignmentWithSolution.ipynb) - 探究IT稳定性问题是否由容量限制引起。\n* [共享单车](https:\u002F\u002Fgithub.com\u002Fchrisjcc\u002FDataInsight\u002Fblob\u002Fmaster\u002FDataChallenge\u002FBikeShare_Challenge.ipynb) - 结合特征选择，使用XGBRegressor、RandomForestRegressor和GradientBoostingRegressor。\n* [航空公司机队细分](http:\u002F\u002Fhtmlpreview.github.io\u002F?https:\u002F\u002Fgithub.com\u002Fatul-shukla-INSEAD\u002FGroupProjectBDA\u002Fblob\u002Fmaster\u002FGroupProject.html) - 达美航空分析。\n* [爱彼迎](http:\u002F\u002Finseaddataanalytics.github.io\u002FINSEADAnalytics\u002Fgroupprojects\u002FAirbnbReport2016Jan.html) - 爱彼迎预订数据分析。\n\n\n\u003Ca name=\"operations-predict\">\u003C\u002Fa>\n#### 预测管理\n* [纠纷预测](https:\u002F\u002Fgithub.com\u002Fzhanghaizhen\u002FFinancial-Service-Complaint-Management\u002Ftree\u002Fmaster\u002Fipynb) - 金融服务投诉管理。\n* [航班延误预测](https:\u002F\u002Fgithub.com\u002Fcavaunpeu\u002Fflight-delays\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fflight-prediction.ipynb) - 利用Keras中的变分自编码器进行迁移学习，以预测航班延误。\n* [电力故障预测](https:\u002F\u002Fgithub.com\u002Fsusano0\u002FElectric-Fault-Prediction\u002Fblob\u002Fmaster\u002FFault_pred.ipynb) - 通过应用简单的机器学习算法来预测电网站点的跳闸事件。\n* [R语言中的热度预测](https:\u002F\u002Fgithub.com\u002Fs-mishra\u002Ffeaturedriven-hawkes\u002Fblob\u002Fmaster\u002Fcode\u002Fmarked_hawkes_point_process.ipynb) - 标记霍克斯点过程。","# business-machine-learning 快速上手指南\n\n`business-machine-learning` 并非一个单一的 Python 包，而是一个由 Sov.ai 维护的**商业机器学习与数据科学应用资源库**。它汇集了会计、客户分析、员工管理、法律合规及运营优化等多个领域的开源项目、代码示例和研究论文链接。\n\n本指南将帮助你快速浏览该仓库，找到适合你业务场景的代码模板并运行示例。\n\n## 环境准备\n\n由于仓库中包含多个独立的子项目（涵盖会计异常检测、客户生命周期价值预测、文本情感分析等），不同项目依赖的技术栈略有差异。大多数核心示例基于 **Python** 生态。\n\n### 系统要求\n- **操作系统**: Windows, macOS, 或 Linux\n- **Python 版本**: 推荐 Python 3.8 及以上版本\n- **包管理器**: `pip` 或 `conda`\n\n### 前置依赖\n在克隆仓库前，建议安装以下通用数据科学库，它们覆盖了仓库中 80% 的示例需求：\n```bash\npip install pandas numpy scikit-learn matplotlib seaborn jupyterlab\n```\n\n针对特定高级项目（如深度学习或 NLP），可能需要额外安装：\n```bash\npip install tensorflow torch transformers nltk spacy\n```\n\n> **💡 国内加速建议**\n> 推荐使用国内镜像源加速安装过程：\n> ```bash\n> pip install -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple pandas numpy scikit-learn jupyterlab\n> ```\n\n## 安装步骤\n\n该项目是一个代码合集，无需全局安装。请通过以下步骤获取本地副本并配置特定项目的运行环境。\n\n### 1. 克隆仓库\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Ffirmai\u002Fbusiness-machine-learning.git\ncd business-machine-learning\n```\n\n### 2. 选择并进入具体项目目录\n仓库按业务部门分类（如 `Accounting`, `Customer`, `Operations` 等）。你需要根据需求进入具体的子项目链接对应的代码库。\n\n**示例：以客户细分（Customer Segmentation）为例**\nREADME 中列出了多个电商客户细分的 Notebook 链接。假设你选择了一个基于电商数据的案例，通常这些代码托管在外部链接或子目录中。如果代码在当前仓库的子文件夹中：\n\n```bash\n# 假设我们关注 Customer 目录下的某个具体分析脚本\ncd Customer\u002FSegmentation\n```\n\n*注意：由于本项目主要作为“索引”指向各个独立的 GitHub 项目（如 `jalajthanaki\u002FCustomer_segmentation`），你可能需要单独克隆感兴趣的具体项目仓库。*\n\n**通用操作模式（针对链接到的外部项目）：**\n1. 在 README 的 \"Table of Contents\" 中找到感兴趣的项目（例如：*E-commerce customer segmentation*）。\n2. 点击链接跳转至具体仓库。\n3. 克隆该具体仓库：\n   ```bash\n   git clone \u003C具体项目的 GitHub URL>\n   cd \u003C具体项目目录>\n   ```\n4. 安装该项目特定的依赖（通常包含 `requirements.txt`）：\n   ```bash\n   pip install -r requirements.txt\n   # 或使用国内源\n   pip install -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple -r requirements.txt\n   ```\n\n## 基本使用\n\n本仓库的核心价值在于提供**可运行的 Jupyter Notebook (`.ipynb`) 脚本**。以下以“客户生命周期价值 (CLV) 分析”为例，展示如何运行一个简单的商业分析模型。\n\n### 场景：计算客户生命周期价值 (CLV)\n参考 README 中 `Customer` -> `Lifetime Value` 部分的 *Pareto\u002FNBD Model* 示例。\n\n#### 1. 启动 Jupyter Lab\n在项目目录下启动交互式开发环境：\n```bash\njupyter lab\n```\n\n#### 2. 打开示例 Notebook\n在浏览器中找到对应的 `.ipynb` 文件（例如 `CLV Analysis.ipynb`）。\n\n#### 3. 运行核心代码逻辑\n典型的商业机器学习流程如下（代码逻辑还原自常见示例）：\n\n```python\nimport pandas as pd\nfrom lifetimes import ParetoNBDModel  # 需先安装: pip install lifetimes\n\n# 1. 加载交易数据 (示例数据结构)\n# 数据应包含：用户 ID, 购买日期, 金额\ndata = pd.read_csv('transaction_data.csv')\n\n# 2. 数据预处理 (转换为 RFM 格式: 频率，最近一次购买时间，总时长)\n# 使用 lifetimes 库的工具函数进行转换\nfrom lifetimes.utils import summary_data_from_transaction_data\n\nsummary = summary_data_from_transaction_data(\n    data, \n    'customer_id', \n    'date', \n    monetary_value_col='amount'\n)\n\n# 3. 训练 Pareto\u002FNBD 模型\npnbd = ParetoNBDModel()\npnbd.fit(summary['frequency'], summary['recency'], summary['T'])\n\n# 4. 预测未来购买概率\n# 预测下一个时间单位的预期购买次数\nsummary['predicted_purchases'] = pnbd.expected_number_of_purchases_up_to_time(\n    t=1, \n    frequency=summary['frequency'], \n    recency=summary['recency'], \n    T=summary['T']\n)\n\nprint(summary[['predicted_purchases']].head())\n```\n\n### 探索其他领域\n你可以按照同样的方式探索其他板块：\n- **Accounting (会计)**: 运行 `Financial Statement Anomalies` 脚本来检测财务报表异常。\n- **Operations (运营)**: 使用 `Failures and Anomalies` 中的代码进行设备故障预测。\n- **Legal (法律)**: 尝试 NLP 工具进行合同条款的政策合规性检查。\n\n> **提示**: 每个子项目通常都有独立的 `README` 或 Notebook 内的 Markdown 说明，请在运行前仔细阅读该具体项目的参数配置要求。","某中型量化对冲基金的数据科学团队正试图利用非传统数据源（如开发者活动日志和卫星图像）来构建新的市场预测模型，以寻找超额收益机会。\n\n### 没有 business-machine-learning 时\n- **资源分散且检索困难**：团队成员需要花费数周时间在 GitHub、arXiv 和各大学术博客中手动筛选与会计、客户流失或员工绩效相关的零散案例，效率极低。\n- **缺乏经过验证的实战路径**：由于找不到针对特定业务场景（如利用 GitHub 日志预测趋势）的成熟应用清单，团队只能从零开始摸索，面临极高的试错成本和失败风险。\n- **难以对接顶尖研究网络**：团队孤立无援，无法接触到来自牛津、剑桥等顶尖机构的研究方法论，也错失了与拥有前沿算法经验的专家协作的机会。\n- **数据清洗方案缺失**：面对金融数据中常见的缺失值问题，缺乏系统性的插补技术参考，导致模型鲁棒性不足，预测结果波动大。\n\n### 使用 business-machine-learning 后\n- **一站式获取 curated 清单**：团队直接利用 business-machine-learning 中按部门（如运营、法律、客户）分类的精选应用列表，迅速锁定了多个可落地的建模方向。\n- **复用成熟的项目范式**：参考列表中关于\"GitHub 日志预测建模”和“卫星数据分析”的具体案例，团队快速复现了基础框架，将研发周期从数月缩短至几周。\n- **融入高端科研生态**：通过该工具关联到的 Sov.ai 平台，团队成功与具有顶级对冲基金背景的博士研究员建立合作，引入了更先进的特征工程技巧。\n- **提升模型稳健性**：借鉴列表中推荐的前沿数据插补技术，有效解决了不完整数据带来的噪声问题，显著提升了投资策略的准确性和稳定性。\n\nbusiness-machine-learning 通过将分散的商业机器学习应用系统化，帮助金融团队大幅降低了从理论探索到实战落地的门槛与时间成本。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ffirmai_business-machine-learning_a6ce87e0.png","firmai","Derek Snow","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Ffirmai_d5c51551.jpg",null,"NYU | Sov.ai ","d.snow@nyu.edu","dereknow","sov.ai","https:\u002F\u002Fgithub.com\u002Ffirmai",[82],{"name":83,"color":84,"percentage":85},"Jupyter Notebook","#DA5B0B",100,813,224,"2026-03-17T14:20:53","CC0-1.0",5,"","未说明",{"notes":94,"python":92,"dependencies":95},"该仓库并非单一的可执行 AI 工具，而是一个商业机器学习与数据科学应用的项目列表和资源索引。README 主要介绍了 Sov.ai 的研究合作机会，并按部门（会计、客户、员工等）分类列出了多个指向外部 GitHub 仓库、Jupyter Notebook 教程或数据集的链接。因此，文中未包含统一的运行环境、依赖库版本或硬件需求说明。具体的环境要求需参考列表中各个独立子项目的文档。",[],[97,16,14],"其他",[64,99,100,101,102,103,104],"practical-machine-learning","jupyter-notebook","jupyter","example","datascience","machine-learning","2026-03-27T02:49:30.150509","2026-04-06T18:54:36.072300",[],[]]