This article summarizes the author's participation in the Kaggle Jane Street Market Prediction competition, detailing the anonymous feature dataset, utility‑score metric, data preprocessing, the combined AE‑MLP and XGBoost modeling approach, threshold tuning, experimental findings, and references for further study.
This article describes how the Beaver search engine’s many performance‑related configuration parameters can be automatically tuned using machine‑learning techniques from OtterTune and AutoTiKV, detailing the background research, Gaussian Process regression model, Bayesian optimization process, implementation steps, test results, and future improvements.
This tutorial explains the concept of facial recognition, describes how it works, and provides step‑by‑step instructions and code examples for implementing face detection and identification using OpenCV and Python, including installation, basic image handling, and a complete sample script.
This article demonstrates how to install the Python face_recognition library, locate faces in images, crop and save them, encode faces into 128‑dimensional vectors, compare faces, detect facial landmarks, apply virtual makeup, and assemble these steps into a simple custom face‑recognition application.
The article explains how eBay protects its global payment system using a graph‑neural‑network driven risk management framework called xFraud, which combines heterogeneous graph sampling, node‑type encoding, attention mechanisms and dynamic‑graph extensions to detect and explain both individual and organized fraud patterns in real‑time.
Intelligent edge computing extends traditional edge architectures by embedding AI-driven data analysis and machine learning directly at the network edge, enabling faster, low‑latency decisions for use cases like smart cameras and autonomous vehicles, and is rapidly becoming a core strategy for modern enterprises.
Meituan Search tackles the multi‑business ranking challenge by introducing a quota‑allocation model (MQM) and a series of precise ranking models (MBN) that progressively incorporate sub‑networks, multi‑task learning and transformer‑based behavior sequences, delivering consistent CTR and purchase‑rate gains across food, hotel, travel and other services while outlining future work on feature utilization, sample‑imbalance mitigation and multi‑objective optimization.
This article explains a four‑level methodology for choosing JavaScript‑based machine‑learning tools, demonstrates practical code examples ranging from NLP with nlp.js to deep‑learning with TensorFlow.js, traditional ML with mljs, and statistical computing with stdlib, and shows how to run them entirely in the browser.
This article summarizes key insights from a Tencent technical live stream, detailing how artificial intelligence enhances advertising efficiency and recommendation accuracy, outlines future AI research and application trends, discusses essential hiring criteria for algorithm engineers, and introduces experimental design challenges in two-sided market platforms.
This article examines GitHub Copilot’s AI-driven code suggestions, its underlying Codex model, multilingual support, performance on LeetCode challenges, and the controversy over potential code copying, presenting a detailed classification of observed cases, statistical analysis, and conclusions about the rarity and nature of actual code recitation.
The paper describes Xianyu’s new relevance‑matching pipeline—integrating basic, text‑matching, semantic (BERT‑based dual‑tower), multimodal, and click‑graph features and fusing them with a GBDT model—which boosts search DCG@10 by 6.5 %, query satisfaction by 24 % and click interaction by over 20 % while outlining future enhancements for finer attribute matching and richer structured data.
This article introduces Xiaomi's knowledge graph architecture, presents text‑graph joint representation learning methods, and demonstrates their practical use in entity linking, entity recommendation, and knowledge completion, highlighting model designs, feature engineering, and experimental results.
SQLFlow is an open‑source platform that lets users build and run machine‑learning and deep‑learning models directly from SQL statements, lowering the barrier for business analysts to apply AI by abstracting complex pipelines into familiar database queries.
This article presents a comprehensive overview of Haola's ride‑sharing algorithm ecosystem, covering the machine‑learning platform foundation, the architecture and evolution of the matching recommendation engine, the transaction‑ecosystem governance models, and the intelligent marketing uplift framework, highlighting technical challenges, solutions, and performance gains.
This article presents an extensive glossary of more than fifty big‑data concepts—including Apache projects, data‑analysis methods, storage formats, AI‑related terms, and emerging metrics—providing concise English explanations for each term.
The paper describes Meituan’s common‑sense concept knowledge graph, detailing a multi‑stage construction pipeline—concept, hierarchy, attribute, bridging, and POI/SPU linking—using BERT, XGBoost, and graph neural networks, and demonstrates its deployment in category‑word enrichment, search suggestions, and medical‑beauty tagging, achieving over two million concepts, three million relations, and roughly 90 % accuracy.
This article explains how Douyu uses graph‑based algorithms to detect and mitigate fraudulent streaming traffic, covering the platform's risk‑control scenarios, the overall graph architecture, its evolution, modeling workflow, practical case studies, and the resulting improvements in detection accuracy and interpretability.
This tutorial walks you through creating a front‑end‑only icon recognition system using machine learning, covering background, key concepts, sample generation with canvas, model training via Pipcook, and deploying the TensorFlow.js model for real‑time icon search.
This article explains how Hopsworks’ feature store unifies online low‑latency and offline high‑throughput storage using a dual‑system architecture built on RonDB, detailing its API, metadata handling, ingestion pipeline, benchmarks, and how it simplifies production machine‑learning feature access.
This article systematically introduces the architecture, iterative improvements, modeling techniques, and practical applications of Meituan's food knowledge graph, covering category taxonomy, standard dish names, basic and thematic attributes, health‑meal detection, dish entity alignment, and downstream recommendation and search use cases.
This notebook provides ready‑to‑use Python code templates for ten common machine‑learning algorithms—including linear regression, logistic regression, decision trees, Naïve Bayes, SVM, K‑Nearest Neighbors, K‑Means, Random Forest, PCA, and Gradient Boosting—showing how to import, train, evaluate, and predict with scikit‑learn.
This article presents the technical exploration of splicing recall in flight ticket search at Alibaba's Fliggy, detailing the background, challenges, constrained routing and machine‑learning algorithms, the four‑step solution pipeline, and the resulting improvements in ticket availability and conversion rates.
This article introduces the third edition of Stanford’s Data Mining Tutorial, highlighting its panoramic roadmap of data‑mining techniques for massive datasets, core features, comprehensive topic coverage, target audience, and supplementary resources while noting its popularity among students and professionals.
This article provides a comprehensive overview of machine learning, explaining core AI terminology, distinguishing statistics, statistical learning, and machine learning, outlining a three‑part learning roadmap covering mathematical foundations, algorithms, and Python programming practice, and offering curated resources for building a solid knowledge system.
This article provides a comprehensive overview of user profiling, covering its definition, the five‑dimensional framework (goal, method, organization, standards, validation), various tag classifications, tag‑system architecture, modeling techniques, practical applications such as precise marketing and product innovation, and a step‑by‑step guide for building a profiling system using big‑data and AI methods.
The Meituan team introduces DPIN, a three‑module deep network that jointly models ads and their positions to mitigate position bias in CTR prediction, achieving up to 2.98% AUC improvement, 2.25% higher CTR and 2.15% RPM gains while keeping latency modest, and is applicable to broader ranking tasks.
This article outlines a four‑layer methodology for selecting JavaScript‑based machine‑learning tools—ranging from domain‑specific NLP libraries to deep‑learning, classic ML, and math‑statistical packages—providing code examples, installation commands, and practical tips for front‑end developers.
Edge‑cloud collaborative machine learning faces data latency, cost, compression, privacy, and heterogeneity challenges, prompting a shift from closed learning to lifelong learning that leverages cloud‑side knowledge bases and edge‑side incremental updates, as demonstrated by the Sedna platform’s thermal comfort prediction case study.
This article details how 58.com’s AI Lab applied machine‑learning, recommendation, search, speech and NLP technologies to transform its CRM system into an intelligent opportunity distribution platform and sales voice assistant, describing the underlying models, the "Michigan" workflow, AB‑testing results and future AI‑driven enhancements.
This article introduces JD.com's 9N GRAPH industrialization framework for large‑scale graph algorithms in advertising, covering the challenges of e‑commerce recommendation, the end‑to‑end solution architecture, the BVSHG multi‑behavior heterogeneous GNN model, training pipelines, and observed business impact.
This tutorial walks through the complete process of building a linear regression model in Python from loading a housing price dataset, normalizing features, defining hypothesis, cost and gradient‑descent functions, visualising data and cost convergence, and testing predictions, with full source code provided.
This article explains the origins, definitions, and evolving relationship between artificial intelligence and machine learning, highlighting their historical milestones, core concepts, and how modern applications like deep learning, neural networks, and recommendation systems illustrate their intertwined development.
The article explains how enterprises must organize increasingly complex data—from structured sources to social media and IoT—to unlock AI value, describing data lakes, hybrid cloud strategies, and the concept of an AI assembly line that integrates tools, containers, and cross‑functional teams for scalable machine‑learning deployment.
The Meituan Takeaway Food Knowledge Graph iteratively builds a hierarchical tag taxonomy, standardizes dish names, extracts basic and theme attributes, aligns online‑offline entities using CNN‑CRF, BERT and hybrid models, and powers combo, interactive and search recommendations while planning scene‑specific tags and graph‑based personalization.
The article reviews Anthony J. Algmin’s reflections on past data‑architecture predictions, current hot topics such as cloud, AI/ML, data governance, and real‑time analytics, and forecasts future trends including metadata management, blockchain, and the evolving role of data architects within enterprises.
This article explains the theoretical foundations of CTR modeling, why click‑through rates are intrinsically unpredictable at the micro level, the simplifying assumptions that make binary classification feasible, and how evaluation metrics like AUC, contradictory samples, theoretical AUC bounds, and calibration affect model performance.
This article explains how causal inference techniques are applied to model the uplift effect of ad placement in Alibaba's Fliggy search CPC advertising, transforming budget allocation into a multi‑objective optimization problem and describing practical control methods, feature engineering, sample re‑sampling, model designs, uplift evaluation, and future research directions.
This article shares personal experiences and insights from using Tencent's TI-ONE machine learning platform in the 2020 Tencent Advertising Algorithm Competition, covering platform features, development modes, resource management, and lessons learned for future participants.
This article outlines eight practical big‑data analytics use cases—from product recommendation and pricing to churn prediction—and introduces fundamental machine‑learning algorithms such as linear regression, decision trees, SVM, and random forests that power these applications.
The article follows three Didi Ride‑Hailing engineers who joined in 2016 as fresh PhDs, detailing how they leveraged machine‑learning, dynamic dispatch, and product‑line automation to rise from junior developers to technical leaders, highlighting the blend of hard coding, soft communication, rapid‑learning culture, and the team’s current senior‑role hiring drive.
This article provides a comprehensive summary of machine learning model evaluation metrics, covering accuracy, precision, recall, F1, RMSE, ROC/AUC, KS test, and scoring cards, with explanations, formulas, code examples, and practical considerations for model performance assessment.
This article provides a comprehensive overview of AIOps, covering its definition, evolution from manual to AI-driven operations, core capabilities, and real-world applications in capacity prediction, anomaly detection, and alarm merging, illustrated with case studies from a food‑retail giant and internal logistics.
The article reviews the evolution of iQIYI's short‑video recommendation recall pipeline, detailing multi‑interest recall methods such as clustering‑based recall, MOE‑based recall, single‑activation multi‑interest networks, regularization strategies, dynamic capacity handling, and multimodal extensions, and discusses their impact on recommendation performance.
The article details how iQIYI doubled online inference speed and cut p99 latency by over 50% in TensorFlow‑based CTR recommendation models by replacing costly string‑based integer hashing, removing redundant dense‑sparse conversions, and deduplicating user features for efficient broadcasting, demonstrating that modest Feature Column tweaks can yield major production gains.
This article introduces the Continuous Causal Forest, a novel uplift modeling approach that expands binary treatment effect estimation to handle multivariate and continuous treatment variables, demonstrates its construction, evaluates its performance on ride‑hailing pricing strategies, and discusses its advantages, limitations, and future research directions.
The 2021 Tencent Advertising Algorithm Competition featured live streams on May 10-12, 2021, with experts discussing the competition's technical aspects and practical applications of the Angel distributed machine learning framework.
This article explains how to use Python's NumPy library to translate common mathematical formulas—such as powers, roots, absolute values, vector and matrix operations—into concise, executable code, covering setup, basic operations, and practical examples for data analysis and machine learning.
The article explains how JD’s retail user growth team built a data‑driven, AI‑powered “machine” that automates user insight, operation planning, conflict resolution, external touchpoint, and real‑time strategy engines to achieve precise, large‑scale user acquisition and retention.
The article reviews the most important additions in scikit‑learn 0.24, including faster hyper‑parameter search methods, ICE plots, histogram‑based boosting improvements, new feature‑selection tools, polynomial‑feature approximations, a semi‑supervised classifier, MAPE metric, enhanced OneHotEncoder and OrdinalEncoder handling, and a more flexible RFE interface.
This article presents a comprehensive overview of sentiment analysis in user‑generated content, detailing document‑, sentence‑, and aspect‑level tasks, defining the Aspect Sentiment Triplet Extraction problem for e‑commerce reviews, describing a three‑stage pipeline with pre‑training, multi‑domain modeling and attribute normalization, and reporting significant business improvements such as 400% CTR lift, while also discussing data imbalance, annotation scarcity, and future research directions.
This guide explains why Python is essential, presents a step‑by‑step learning roadmap covering beginner basics, backend web development, web crawling, data analysis, and machine learning, and provides curated resources and project links to help learners progress efficiently.
This tutorial walks through a practical case of predicting employee attrition, demonstrating how to assess and compare machine‑learning models using ROC AUC, confusion matrices, precision‑recall trade‑offs, and the Evidently library to generate performance dashboards, helping choose the best model for production.
This article reviews the limitations of traditional DMP and AI‑driven intelligent targeting in e‑commerce, introduces JD.com's Smart DMP framework that combines merchant intent with high‑relevance modeling, and presents experimental results showing over 15% CTR improvement and widespread merchant adoption.
This article systematically introduces Alibaba's advertising intelligence platform, covering the evolution from basic CPM/CPC models to advanced OCPC/OCPM, budget‑constrained bidding, multi‑constraint bidding, sequence‑based long‑term value bidding, multi‑channel allocation, and the AI‑driven Smart Bidding product, highlighting algorithmic foundations, practical implementations, and performance gains.
The article presents a comprehensive guide to Python's built‑in functions, standard libraries, and third‑party packages across file I/O, web scraping, databases, data cleaning, statistical analysis, machine learning, visualization, and more, rating each with stars and offering a free e‑book collection for readers.
Python has become the go‑to language for AI and machine learning across startups and enterprises because of its rapid prototyping, flexible syntax, readability, extensive libraries like NumPy, SciPy, scikit‑learn, Pandas, Keras, and powerful visualization tools, making development faster, scalable, and easier to maintain.
This article analyzes iQIYI’s evolution of multi‑interest recall techniques—from clustering‑based PinnerSage to MOE and single‑activation models—showing how extracting multiple user interests improves recall diversity, mitigates filter bubbles, and boosts key performance metrics in short‑video recommendation.
The 17‑minute video replay of Gaode's Technology Open Day showcases Cui Hengbin’s deep‑dive into AI‑driven route planning, covering pre‑trip, in‑trip, and arrival phases, road‑condition prediction, ETA estimation, dynamic traffic mining, and references to award‑winning research papers.
This article presents Didi's comprehensive governance algorithm system for O2O platforms, detailing business background, technical challenges, a three‑stage algorithmic framework, model innovations such as small‑sample learning, multi‑task and transfer learning, and extensive feature engineering including multimodal and streaming features.
This article introduces a personalized re‑ranking model for recommendation systems, explaining the limitations of traditional point‑wise ranking, describing the PRM architecture with input, encoding, and output layers using multi‑head attention and pre‑trained personalization features, and presenting experimental results and future extensions.
The 3rd International Workshop on Deep Learning Practice for High-Dimensional Sparse Data (DLP‑KDD 2021) invites submissions on deep‑learning systems, data representation, and user modeling for large‑scale sparse data, with a submission deadline of May 10 2021 and results announced on June 10 2021.
The article describes a unified global feature pool architecture that standardizes offline and real‑time feature production, management, and service layers using Hive, Spark, Flink, Kafka, MySQL, and Hologres to break data silos, improve algorithm development efficiency, and boost growth business performance.
The ML‑Summit 2021 in Beijing featured two AI‑focused talks—one on applying AI to CRM for boosting enterprise performance and another on systematic recommendation recall and ranking optimization—each presented by senior engineers from 58.com, with detailed abstracts and speaker biographies.
This article compiles and summarizes fifteen influential survey papers on Graph Neural Networks, covering their models, applications, datasets, benchmarks, challenges, and future directions, while providing links to the original PDFs and highlighting distinctions between small and large-scale graph learning.
This article surveys recent advances in graph neural network explainability, systematically categorizing instance‑level and model‑level methods, reviewing datasets, evaluation metrics, and proposing new benchmark graph datasets for interpretable GNN research, and highlighting future research directions.
Learned Index Structures replace conventional B‑Tree and Bloom filter indexes with hierarchical machine‑learning models that predict key positions directly, reducing search overhead, but require careful model selection, training, and handling of updates, making them a promising yet still experimental alternative to traditional database indexing techniques.
This article presents a graph‑neural‑network approach to predict user influence, cluster users with distributed Louvain methods, and conduct network‑aware A/B experiments that accurately evaluate large‑scale marketing campaigns despite strong network effects.
This article presents the practical deployment of OCR technology within 58’s information‑security workflows, focusing on layout‑analysis techniques for document and credential recognition, detailing rule‑based, template‑matching, object‑detection, and image‑segmentation methods, their implementation steps, experimental results, advantages, limitations, and future directions.
This article explains the architecture and ranking process of modern recommendation systems, covering the two-stage pipeline of candidate generation and ranking, the evolution from rule‑based methods to logistic regression, GBDT, wide‑and‑deep, and deep learning models, and discusses challenges such as feature non‑linearity, multi‑objective optimization, and the need for post‑ranking interventions.
This article presents a comprehensive model quality assurance framework for click‑through‑rate (CTR) prediction, detailing the challenges of data and logic inconsistency, defining consistency goals, describing a full‑stack verification pipeline—including online data capture, offline sample alignment, multi‑stage q‑value comparison, and automated reporting—and sharing practical deployment experiences and results.
This article presents Ant Financial's intelligent middle platform, detailing AI use cases such as risk control, wealth management, lending, marketing, insurance, and customer service, alongside the AI capability map, data foundation architecture, annotation workflows, security measures, and the overall impact on fintech innovation.
This article introduces the three main categories of machine learning—supervised, unsupervised, and reinforcement learning—detailing common algorithms such as Linear Regression, Logistic Regression, Naive Bayes, K‑Nearest Neighbors, Decision Trees, Random Forests, SVM, K‑Means, and PCA, and provides concise Python code examples using scikit‑learn for each.
This tutorial explains why and how to reduce the number of features in the Kaggle Mushroom Classification dataset with Python, covering preprocessing, various feature‑selection methods (filter, wrapper, embedded), code examples, model training, performance impact, and visualisation of results.
This article presents the design and implementation of interactive recommendation and travel‑theme recommendation in Alibaba's Fliggy app, covering background, user demand classification, real‑time interest capture, various recall strategies, ranking models, multi‑task learning, and engineering tricks to improve CTR and user experience.
This article introduces a quality scoring model that leverages structured development and testing data to objectively assess project risk, automate test grading, and enable data‑driven decisions for test execution and release, thereby improving delivery efficiency and reducing manual evaluation errors.
This article details how Toutiao's senior algorithm architect designs a transparent recommendation system, covering system overview, three-dimensional feature modeling, real-time training pipelines, recall strategies, content analysis, user tagging, evaluation methods, and content safety measures.
TAL Education (好未来) has opened a 587‑hour bilingual Chinese‑English speech dataset from classroom teaching, one of the largest open educational corpora, aiming to fill the data scarcity in mixed‑language speech recognition research and support AI model development.
This article analyzes the rapid growth of O2O home‑delivery, examines the challenges of delivery fulfillment, compares third‑party and self‑built rider models, and presents hybrid, batch‑ordering, and AI‑driven optimization strategies to reduce costs and boost efficiency.
This article reviews the 2020 achievements of Alibaba.com’s ICBU algorithm team, explaining the evolving role of algorithm engineers, the fundamentals of e‑commerce algorithms, the team’s three‑pillar workflow of Understanding, Growth, and Matching, and the technical breakthroughs that drove business impact and future directions.
This article details Kuaishou's multi‑objective ranking pipeline for short‑video recommendation, covering manual score fusion, GBDT ensemble, Learn‑to‑Rank, online auto‑tuning, ensemble sorting, reinforcement‑learning rerank, and on‑device rerank, and reports their impact on DAU, watch time and user interaction.
This article introduces the fundamentals of graph neural networks, explains their expressive power for relational risk identification, and details Ant Group's innovations—including the GeniePath architecture and a bandit‑based sampling optimizer—that achieve superior performance on benchmark datasets.
This article explains how Meituan’s delivery business has transitioned from data online to AI‑driven decision making by building a comprehensive, one‑stop machine learning platform that includes model management, data graph, feature store, AB testing, and a machine‑learning definition language to accelerate algorithm iteration and reduce operational costs.
The Xianyu pricing guidance system blends new‑product market values with depreciation factors derived from usage, condition and category attributes—extracted via real‑time text mining and image analysis—to recommend dynamic price ranges adjusted for supply‑demand and seller urgency, currently covering 60% of listings with over 65% overall accuracy.
The article outlines a three‑dimensional framework—technical, business, and soft‑skill competencies—that algorithm engineers need to master in order to successfully deliver machine‑learning solutions in production environments, offering practical advice on data handling, model evaluation, stakeholder communication, and personal development.
This article introduces Pipcook, a machine‑learning framework designed for front‑end developers, explains its architecture and integration with TensorFlow.js, Boa, and Node.js, and discusses how it lowers the barrier to building intelligent front‑end applications through pipelines, plugins, and cloud‑native deployment.
This article reviews the development history, technical routes, and recent breakthroughs of pre‑ranking (coarse ranking) in large‑scale advertising systems, focusing on Alibaba's COLD (Computing‑power‑cost‑aware Online and Lightweight Deep) framework, its model design, engineering optimizations, experimental results, and future research directions.
The article examines how data teams can select the appropriate Ops framework—DataOps, MLOps, or AIOps—by comparing their origins, principles, responsibilities, and tooling, and stresses that cultural principles outweigh technology choices for efficient delivery of data and machine‑learning products.
This article explores the intersection of natural language processing and operations, outlines common text‑handling challenges, and presents three concrete AIOps case studies—log Q&A, anomaly detection, and ticket recommendation—while reflecting on a closed‑loop AI workflow and future research directions.
This article explains how Fliggy combines interactive recommendation with travel‑theme recommendation, detailing the underlying algorithms, user‑demand classification, real‑time interest capture, recall strategies, multi‑task learning for CTR prediction, and engineering tricks that improve personalization and click‑through rates.
This article explains on‑device machine learning concepts, compares front‑end inference engines such as TensorFlow.js, ONNX.js and WebDNN across CPU, WASM and WebGL, and presents practical optimization techniques like vectorization, memory layout, graph fusion and mixed‑precision to boost performance for real‑time applications.
This article systematically reviews the evolution of click‑through‑rate (CTR) prediction models—from early distributed linear models like logistic regression, through automated feature engineering with GBDT+LR, various factorization‑machine variants, embedding‑MLP shallow modifications, dual‑tower combinations, and advanced explicit feature‑cross networks—highlighting each model’s structure, advantages, limitations, and comparative insights.
This article provides a detailed introduction to machine learning, covering its definition, learning modes such as supervised, unsupervised and reinforcement learning, shallow versus deep learning, the full industry chain from AI chips to cloud and big‑data services, and the major open‑source frameworks and platforms driving the field.
This article presents Hulu's multi‑layered approach to content understanding and embedding, describing tag‑based graph embeddings, metadata‑BERT enhancements, multimodal video/audio feature aggregation, and various applications such as similarity search, ranking, cold‑start retrieval, and collection modeling, while also discussing current limitations and open research questions.
The article explains the origins and differences of DevOps, DataOps, MLOps and AIOps, outlines their shared seven principles, and provides practical guidance on adopting the right X‑Ops culture to accelerate data‑driven and machine‑learning‑powered software delivery.
This article provides a visual, step‑by‑step introduction to NumPy’s core concepts—vectors, matrices, higher‑dimensional arrays, creation, indexing, arithmetic, broadcasting, and common functions—helping developers and researchers understand how the library works and apply it efficiently in Python data‑science workflows.
Meituan’s unmanned delivery team secured first place in the Generalizability track and second in the Regular track of the NeurIPS 2020 INTERPRET trajectory‑prediction challenge by employing a mixed‑attention graph‑transformer with dual‑channel GRU and adaptive map processing, achieving ADEs of 0.5339 m and 0.1912 m respectively.
This article explains how feature engineering transforms raw data into numerical representations for machine‑learning models, covering mapping of numeric and categorical values, one‑hot and multi‑hot encoding, sparse representations, scaling, handling outliers, binning, data quality checks, and feature interactions to capture non‑linear relationships.
This article describes how 58.com’s AI Lab integrated machine‑learning and recommendation techniques into its CRM system, redesigning sales workflows, introducing the “Michigan” model, and deploying XGBoost and MMoE models to boost key metrics such as transfer rate and 60‑second effective call rate, achieving significant performance gains.
Time series databases let you retain full historical records, enabling analysis, visualization, machine learning and automation across domains like finance, weather and IoT, and the article explains why they’re essential, how they differ from traditional databases, and how to start using them.
This article presents a comprehensive overview of using graph representation learning and clustering, particularly GraphSAGE and its optimizations, to identify and mitigate black‑market (malicious) accounts in virtual networks, discussing background, objectives, challenges such as isolation and heterogeneity, and evaluation results.
