This article explains why meticulous feature engineering remains crucial in recommendation and search advertising, outlines what constitutes good features, describes common transformation techniques such as scaling, binning, and encoding, and provides practical examples and Q&A for practitioners.
Yanxuan built a Kubeflow‑based machine‑learning platform that unifies data preprocessing, feature engineering, model training, validation, and deployment, using Smart‑jobs, Smart‑Infer, Smart‑backend, Airflow pipelines, Jupyter notebooks, and Istio‑enhanced inference services to boost algorithm engineers’ efficiency and integrate with Kubernetes, HDFS, and Hive.
The talk by Tang Gengyang from Citic Baixin Bank details the challenges faced in risk feature engineering, presents two solution frameworks (1.0 and 2.0) for accelerating deployment, improving reuse, handling offline/online consistency, and outlines future enhancements for a more efficient, automated feature pipeline.
This presentation details 360ShuKe's risk control system architecture, covering its layered design, credit data lifecycle management, real‑time indicator computation, feature platform evolution, and solutions to challenges such as data loss, rapid model iteration, and feature drift.
This article presents Huace Data Science's practical approaches to digital supply‑chain finance for SMEs, detailing challenges of cross‑industry data, the SME engine for authentic business assessment, graph‑based fraud detection, and quantum‑inspired feature‑engineering methods that enhance credit‑risk models.
This article explains the concept and execution methods of rule engines, describes the forward‑link inference approach used by ZhiZhuan’s risk‑control engine, and details the evolution from hard‑coded rules to feature‑engineered expressions using the Aviator expression engine, highlighting configuration, data sources, and practical examples.
The article details Meituan's engineering journey from small DNNs to hundred‑gigabyte deep learning models for food‑delivery ads, analyzing online latency and offline efficiency challenges and presenting distributed storage, CPU/GPU acceleration, OpenVINO, TensorRT, CodeGen, and data‑pipeline optimizations that dramatically improve throughput, memory usage, and sample‑building speed.
This article details Meituan’s implementation of on‑device deep learning models for search re‑ranking, covering the motivations for edge intelligence, feature engineering, feedback sequence modeling, model architecture, deployment optimizations, experimental results, and future directions, offering practical insights for developers building large‑scale AI on mobile.
This article introduces the DolphinScheduler OpenMLDB Task, explains how it integrates OpenMLDB's feature platform into DolphinScheduler workflows to create a complete MLOps pipeline, and provides a step‑by‑step demonstration using the TalkingData ad‑fraud detection dataset from Kaggle.
This article explains how a multimodal deep learning model combines static and temporal game data to predict frame rates, helping identify performance bottlenecks and improve client smoothness through feature fusion, data pipelines, and real‑time inference in modern games.
This article details Amap's practical experience with Learning to Rank for search suggestions, covering application scenarios, data pipeline construction, feature engineering, model training, loss‑function adjustments, and the resulting performance improvements, while also discussing challenges such as sparse features and click bias.
This article examines the challenges of traditional data feature acquisition and presents Ant Group’s ultra‑real‑time client feature center, detailing its architecture, data collection, streaming and script computation, backflow mechanisms, and monitoring to deliver rich, timely, and easy‑to‑use features for AI models.
Coarse ranking bridges recall and fine ranking by trimming tens of thousands of candidates to a few hundred or thousand using a three‑part framework—sample construction, ordinary and cross‑feature engineering, and evolving deep models—from rule‑based to lightweight MLPs, while employing distillation, feature crossing, pruning, quantization, and bias mitigation to balance accuracy with strict latency constraints.
The article explores the concept of nonlinearity in machine learning, illustrating why tasks like distinguishing cat versus dog or predicting body shape from height and weight are challenging for linear models, and discusses feature engineering, kernel tricks, and periodic activation functions as strategies to introduce nonlinearity and improve model performance.
The article presents Snowball's model feature platform, detailing its motivation, architecture, feature lifecycle management, online engine design, optimization techniques, and the resulting improvements in feature iteration speed, reuse, and system stability for recommendation and search services.
NetEase Yanxuan replaced its cumbersome data‑warehouse with a flexible Delta‑Lake/Iceberg data lake, creating a unified metadata layer and real‑time ingestion pipelines that cut latency from nightly batches to seconds, slashed compute and storage costs, supported diverse business scenarios and machine‑learning feature engineering, and set the stage for broader future expansion.
This article explains how NetEase Yanxuan evolved from a traditional data‑warehouse pipeline to a cloud‑native data‑lake architecture, detailing the business challenges, design choices, technology stack (Delta, Iceberg, Hudi), implementation steps, and the resulting gains in real‑time data access, cost reduction, and feature‑engineering support.
Hello Travel tackles information‑flow advertising challenges by using LightGBM‑based models to predict order conversion, creative performance, and pre‑bid user quality, augmenting sparse data with feature engineering and uplift techniques, while planning future fully automated delivery, richer pre‑screening, and cross‑channel reinforcement‑learning enhancements.
The article provides a thorough guide to ranking in recommendation systems, detailing the pipeline architecture, sample handling challenges, extensive feature engineering categories, the evolution from collaborative filtering to deep and attention‑based models, and key optimization trade‑offs between memorization, generalization, and efficient user‑interest modeling.
This article details the architecture, incremental model training, feature engineering, and deployment strategies that enabled NetEase Cloud Music to achieve real‑time live‑stream recommendation, covering business background, multi‑objective modeling, real‑time feature pipelines, sample attribution, feature admission, and online performance results.
This report details how the Search Recommendation team at 58.com upgraded their deep learning ranking model for recruitment by adding multi-valued and semantic vector features, integrating conversion sequences, employing feature‑crossing techniques, optimizing offline data pipelines, and planning future multi‑scene improvements to boost CTR and relevance.
This article presents a comprehensive, practical methodology for feature engineering that combines brute‑force AutoML‑style dictionary searches, business‑logic‑driven feature creation, and feature‑importance‑guided refinement, illustrating each approach with real Kaggle competition examples and concrete code snippets.
This article outlines Akulaku’s background, explores its business scenarios and challenges, details how OpenMLDB’s unified feature‑engineering platform and consistent technology stack address real‑time and offline data processing needs, presents performance comparisons across use cases, and offers cost‑benefit analysis and future improvement suggestions.
This article explains how a five‑line pandas script can generate frequency‑based encodings for categorical columns, why such encodings help low‑frequency categories, and how they integrate with feature crossing and industry use cases like ad exposure, PV/UV counting, and fraud detection.
This article surveys classic recommendation algorithms—including Logistic Regression, Factorization Machines, Field‑aware FM, Wide & Deep, DeepFM, DCN, and xDeepFM—explaining their principles, feature preprocessing, problem scopes, and industrial applications within personalized recommendation systems.
This article presents a comprehensive overview of constructing an intelligent risk control tool system, detailing its evolution from manual processes to automated platforms, describing the core "three‑piece" suite (model, decision, and feature platforms) along with supporting data and monitoring platforms, and explaining the functions and interactions of each module such as data ingestion, feature engineering, automated modeling, decision flow, and real‑time monitoring.
This article analyzes how often common machine‑learning algorithms such as k‑NN, Naïve Bayes, decision trees, SVM, logistic regression, and neural networks are used in industry, explains their typical scenarios, highlights strengths and weaknesses, and shows how non‑linearity and feature engineering affect their suitability.
This article explains the concept of feature engineering, illustrates it with a height‑weight classification example, compares kernel‑enhanced models to handcrafted features like BMI, discusses its impact on model performance, and highlights practical tips and domain‑specific considerations.
This article introduces CatBoost, outlines its key advantages such as automatic handling of categorical features, symmetric trees, and feature combination, and provides a step‑by‑step Python tutorial—including data preparation, model training, visualization, and feature importance analysis—using a CTR prediction dataset.
This article presents a three‑part technical guide covering the fundamentals of computational advertising and real‑time bidding, detailed offline pCTR model training pipelines with feature engineering, calibration and model structure improvements, and advanced online learning techniques such as parameter freezing, sample replay and knowledge distillation, all aimed at boosting CTR performance and reducing bias in large‑scale ad platforms.
iQIYI’s generic ranking framework unifies feature production, replay, training, and ranking into modular, configurable phases that handle offline and real‑time data, support diverse models, provide automated monitoring, and have been deployed across all platforms, delivering over 20% higher watch time and doubling first‑play videos.
This article presents a comprehensive case study of applying edge AI to video preloading on the Xigua Video platform, detailing scenario analysis, predictive modeling of user behavior, feature engineering, on‑device model inference, dynamic algorithm package deployment, experimental evaluation, and the resulting performance and cost improvements.
This article describes how a financial company leveraged deep‑learning sequence models to automatically extract features from massive time‑series data, improving risk‑assessment models and operational efficiency through a unified framework that includes data preprocessing, embedding, field and item aggregation, and end‑to‑end deployment.
The 58.com AI algorithm competition showcased over 210 teams competing to improve job recommendation click‑through and conversion rates, featured an award ceremony with speeches, highlighted the ten winning teams, and presented detailed solution shares—including tree models, feature‑engineering techniques, and deep‑learning approaches—while offering GPU resources on the WPAI platform for continued participation.
This article explains the architecture and key components of Toutiao’s recommendation system, covering system overview, content analysis, user tagging, evaluation methods, and content safety measures, and discusses practical implementation details such as feature engineering, model training, recall strategies, and online experimentation.
This article details the design, refactoring, performance optimization, reliability monitoring, and commercialization of a financial risk control system, covering rule abstraction, decision workflows, feature engineering, model integration, and the trade‑offs between latency and accuracy in large‑scale production environments.
This article presents a comprehensive overview of Zhihu's recommendation page ranking system, detailing the request flow, ranking evolution from time‑based to deep‑learning models, feature engineering strategies, model architectures such as DNN, DeepFM, DIN, multi‑task learning, and lessons learned for production deployment.
The article describes Baidu’s end-to-end consistency testing framework for advertising click-through-rate models, which uses a five-stream verification pipeline and six implementation phases to compare Q-values across feature extraction, table conversions, and DNN computation, enabling precise detection and localization of data and model inconsistencies in production.
This article presents the design, implementation, and optimization of a deep‑learning‑based ranking system for NetEase News, covering pipeline architecture, feature‑processing enhancements, custom TensorFlow operators, and modular model frameworks such as DCN and DIEN to improve recommendation performance.
This article reviews the evolution of end-to-end deep learning models, describes how they were adapted and optimized for a real‑estate broker churn‑warning scenario, and presents experimental results showing significant improvements in AUC, KS and lift over traditional classifiers.
This article explains how an intelligent risk control system can achieve fully automated, end-to-end model iteration, detailing the multi-layer architecture, sample and feature selection, automated training, evaluation, scoring, deployment, and the efficiency gains compared with manual processes.
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 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.
The paper presents Didi’s three‑layer governance‑algorithm framework for O2O ride‑hailing, addressing high business complexity, limited labeled data, interpretability, and multimodal features through small‑sample, transfer, and multi‑task learning, achieving notable gains in dispute resolution, NPS and CPO while highlighting remaining data and robustness challenges.
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.
This article describes how Ctrip improved hotel recommendation by iterating from logistic regression to GBDT and deep neural networks, designing continuous and discrete features, adopting multi‑task learning with click and conversion signals, and building a large‑scale distributed DNN training and unified feature‑processing framework to boost model accuracy and engineering efficiency.
This article details Meituan Waimai's evolution from a simple feature framework to a sophisticated, configurable platform that handles massive feature production, multi‑task scheduling, dynamic protobuf storage, and a model‑feature description language (MFDL) to enable efficient online retrieval, high‑performance computation, and consistent training‑sample generation for its recommendation, advertising, and search services.
The article describes JD's end‑to‑end online learning pipeline for retail search, covering the background, system architecture, real‑time feature collection, sample stitching, Flink‑based incremental training, parameter updates, and full‑link monitoring to achieve low‑latency, high‑accuracy model serving.
This article details how 58.com tackled the challenges of multi‑business recommendation on its home page by developing a dual‑channel deep interest model, introducing customized feature‑crossing, optimizing training and online performance, and exploring multi‑channel extensions for broader scenario adaptation.
This article describes how 58.com’s rental platform leverages deep learning models such as Wide&Deep, DeepFM, DCN, DIN, and DIEN to improve search ranking, detailing data pipelines, feature engineering, model iteration, multi‑task training, prediction optimizations, and resulting online performance gains.
The AMAP‑TECH competition challenged participants to infer real‑time road conditions from in‑vehicle video, prompting the authors to combine lane‑wise vehicle detection with LightGBM and later an end‑to‑end DenseNet‑GRU model, augment data, ensemble five networks, and achieve a 0.7237 F1 score while outlining future deployment and research directions.
This article presents the end‑to‑end design, feature engineering, model evolution (Wide&Deep, DeepFM, DCN, DIN, DIEN), multi‑task training, and deployment optimizations that 58.com applied to improve search ranking for its rental business, demonstrating significant gains in click‑through and conversion rates.
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.
To tackle the AMAP‑TECH competition’s dynamic road‑condition classification from scarce, imbalanced vehicle‑video frames, the team combined YOLOv5 object detection, ResNeXt101‑based semantic embeddings, and engineered temporal detection statistics, feeding the fused features into a five‑fold LightGBM model that achieved top weighted‑F1 performance.
The article details how 58.com tackled the challenges of multi‑business recommendation by evolving its ranking models from a dual‑channel deep interest architecture to a 1+N multi‑channel deep interest model, incorporating customized feature cross layers, scenario‑adaptation mechanisms, and extensive engineering optimizations that yielded significant CTR and conversion gains.
This article explores the concept of data sensitivity for algorithm engineers, defines its meaning, discusses how to measure it, offers practical steps to develop the skill through data analysis and feature engineering, and reveals the hidden pattern in a label‑prediction example that illustrates its importance.
This talk details the design, data pipeline, feature engineering, model evolution, and operational insights of the 58 Tongzhen home feed recommendation system, covering its architecture, localization strategies, recall and ranking models, online learning, and future directions for AI-driven content delivery in the down‑market.
Xianyu’s premium‑product library combines interpretable, multi‑dimensional metric models built from structured product and user attributes with real‑time and offline pipelines to systematically tag high‑quality items, delivering services via HSF and a message bus, and has driven over 20% click‑through growth and nearly doubled conversion rates.
This article details the design, data pipeline, feature engineering, model development, and iterative optimization of the 58 Tongzhen local feed recommendation system, covering business background, user profiling, recall strategies, ranking models such as XGBoost, XDeepFM, and online learning, and future directions.
This article provides an extensive overview of feature engineering, covering feature understanding, cleaning, construction, selection, transformation, and dimensionality reduction techniques, illustrated with Python code using the Titanic dataset, and offers practical guidelines for improving data quality and model performance in machine learning projects.
This article describes how 58.com applied deep‑learning techniques—including feature engineering, sample construction, model evolution from Wide&Deep to DIN/DIEN and multi‑task learning—and system‑level optimizations to improve CTR/CPM performance in its large‑scale commercial ranking platform.
This article details the end‑to‑end deep‑learning workflow for click‑through‑rate (CTR) prediction in 58.com’s commercial ranking system, covering system architecture, feature engineering, sample construction, model evolution from Wide&Deep to DIN/DIEN, and engineering optimizations that together yielded significant CPM and CVR improvements.
This article explains how large-scale recommendation systems rely on efficient feature engineering, describes the three-layer architecture (offline, stream, online), and details how Spark SQL and the LLVM‑optimized FESQL engine improve performance and ensure offline‑online feature consistency.
This article presents the concepts, architecture, and practical techniques of an online real‑time feature system used in intelligent risk‑control, covering feature definition, time‑window types, calculation functions, distributed processing, low‑latency storage, and operational challenges in high‑concurrency environments.
This tutorial walks through the complete feature‑engineering pipeline for the Alibaba Tianchi “Tmall User Repeat Purchase Prediction” competition, covering data acquisition, memory‑efficient preprocessing, multi‑entity feature construction, statistical aggregations, text vectorisation, embedding generation and stacking‑based model features, all illustrated with Python code and diagrams.
This article provides a comprehensive technical overview of 今日头条's recommendation system, covering its three-dimensional feature model, algorithm choices, real‑time training pipeline, recall strategies, content analysis, user tagging, evaluation methods, and content‑safety mechanisms.
This article shares Weibo’s experience in building and evolving its recommendation algorithms, covering the recommendation scenario, machine learning workflow, feature engineering, model upgrades, large‑scale challenges, deployment via the Weiflow platform, and the capabilities of its machine‑learning infrastructure.
This article details Likee's short‑video recommendation pipeline, covering the evolution of its feature‑engineering framework, the use of Flink for minute‑level statistical and second‑level session features, the integration of automatic statistical features into DNN models, multimodal feature extraction, and the cache‑tensor technique that dramatically improves online inference performance.
Meituan’s AI‑driven search combines the Augur online inference framework—offering stateless, distributed feature operators, transformers, and a DSL for rapid, high‑throughput model scoring—with the Poker platform for model training, versioning, and experimentation, together accelerating iteration, improving performance, and enabling advanced model‑as‑feature ensembles.
This article explains the concept of timeliness in search ranking, defines content and demand side metrics such as half‑life and time sensitivity, describes evaluation criteria, outlines feature extraction and labeling pipelines, and details the multi‑stage modeling, recall, and indexing strategies used to improve timely search results.
This article details how Amap leverages deep‑learning techniques—including temporal and auxiliary feature engineering, multi‑stage RNN models, Wide&Deep architectures, and an Attention‑TCN approach—to accurately identify and handle expired points of interest, improving map freshness and user experience.
NetEase improves its news‑feed recommendation by applying a multi‑stage semantic text understanding pipeline—lexical analysis, hierarchical content tagging, and quality filtering—using two‑level classifiers, LDA‑based topic modeling, multi‑label concept and entity extraction, and dense vector representations to better capture user interests and boost personalization performance.
Feature engineering, which occupies the majority of data scientists' time, is essential for building high‑performing machine‑learning models and involves careful data quality control, diverse construction techniques, rigorous selection, and emerging automation efforts, all of which demand domain expertise and systematic practice.
This article explains why feature engineering consumes most of a data scientist's time, outlines its critical steps—including data observation, cleaning, transformation, selection, and reduction—covers practical issues such as missing‑value handling, data leakage, and feature stability, and discusses both manual and automated approaches for building effective machine‑learning models.
This article reviews recent innovations in CTR prediction for an intelligent marketing platform, covering deep spatio‑temporal networks, deep memory networks, and a feature‑expression‑assisted learning framework, with system architecture details, experimental results, and references to KDD and IJCAI papers.
The project develops a deep‑learning pipeline for Amap’s expired POI detection that integrates two‑year temporal trend features, industry and verification attributes, a variable‑length LSTM, a Wide‑Deep architecture, and a Wide‑Attention Temporal Convolutional Network, achieving higher accuracy and efficiency while outlining future macro‑and micro‑level enhancements.
This article introduces the concept and intuition behind the k-Nearest Neighbor (kNN) classification algorithm, explains its simple and full forms, discusses feature engineering and Euclidean distance calculations, and provides a complete Python implementation with example code.
Meituan’s search ranking system integrates Transformer‑based models across feature engineering, behavior sequence modeling, and re‑ranking, adapting AutoInt‑style embeddings and multi‑stage attention mechanisms to boost QV_CTR and NDCG, while outlining future enhancements with BERT, graph neural networks, and reinforcement learning.
The talk presents Alibaba’s hierarchical video‑category system and a multimodal classification pipeline—leveraging EfficientNet, NeXtVLAD fusion, attention‑dropping augmentation, and MoCo contrastive learning—that together boost cold‑start recall by 43%, improve program classification over 20%, and set the stage for larger models and advanced unsupervised methods.
This article details the evolution of Feizhu's "Guess You Like" ranking system, moving from a linear FTRL model to several end‑to‑end deep learning versions—including PALM, FB‑PALM, and GLA—highlighting technical challenges, architectural changes, and measurable performance gains.
This article describes how 58 Tongzhen leverages AI technologies—including data pipelines, feature engineering, various recall and ranking models, and AB‑testing—to build a personalized feed recommendation system for the down‑market, detailing its overall architecture, data sources, model iterations, performance gains, and future directions.
This article presents a comprehensive overview of the Automobile Home recommendation system, detailing its objectives, architecture, various ranking models from LR to DeepFM, online learning mechanisms, service APIs, feature engineering pipelines, model training platforms, debugging tools, and future optimization directions.
This article presents a comprehensive overview of Phoenix News's AI‑driven feed recommendation system, detailing its business challenges, multi‑stage architecture, deep learning models, feature pipelines, metric trade‑offs, cold‑start solutions, and practical insights for improving user satisfaction and content quality.
This article describes how logistic regression models are built, iterated, and applied to predict student renewal and refund behavior in an online school, detailing data collection, feature engineering, model training, evaluation, and how the predictions are used to recommend interventions for teachers.
This article details the architecture, model evolution, feature processing, online learning, and future optimization plans of AutoHome's recommendation system, covering stages from resource collection to ranking, various models such as LR, XGBoost, FM, DeepFM, and operational practices like AB testing and debugging.
This article describes the end‑to‑end recommendation pipeline of the Province Money Fast Report app, covering business background, data collection, model training and evaluation, the evolution from FM to DeepFM, DIN, DCN, xDeepFM, ESMM and custom networks, as well as serving strategies and practical lessons learned.
This article provides a comprehensive technical overview of Toutiao’s recommendation system, covering its three‑dimensional modeling approach, feature engineering, user‑tag pipelines, real‑time training infrastructure, evaluation methodology, and content‑safety mechanisms.
This article explains how intelligent marketing transforms traditional, labor‑intensive strategies into data‑driven, AI‑powered systems by detailing the multi‑layer architecture, data pipelines, machine‑learning models such as LR and GBDT+LR, and future directions like personalized copy generation and deep‑learning enhancements.
This article provides a comprehensive overview of Toutiao's recommendation system, covering its three-dimensional modeling of content, user, and environment features, various algorithmic approaches, feature extraction, real‑time training pipelines, recall strategies, user‑tag engineering, evaluation methods, and content‑safety measures.
The article details the five‑year evolution of Autohome's recommendation system, covering its overall architecture, the progression of ranking models from LR to DeepFM and online learning, feature engineering pipelines, ranking service APIs, AB testing practices, and future optimization directions.
The article describes a risk‑control pipeline for detecting high‑volume fraudulent game accounts, detailing data collection from game logs, extensive feature engineering and statistical tests, enrichment via a Neo4j knowledge graph, and a hybrid RandomForest‑GBDT model combined with methods to filter personal accounts.
This article details Toutiao’s large‑scale recommendation system, explaining how it models content, user, and environment features, the variety of algorithms and real‑time training pipelines used, feature engineering categories, recall strategies, content analysis, user tagging, evaluation methods, and content‑safety mechanisms.
Meituan’s Turing platform consolidates the entire delivery‑order workflow—from massive data ingestion and feature generation to model training, evaluation, deployment, real‑time prediction, and AB testing—into a single, end‑to‑end system that evolved from a minimal MVP into a fully platformized solution, addressing speed, accuracy, and engineering‑algorithm decoupling while planning deeper deep‑learning integration.
This article reviews the evolution of ranking models in modern recommendation systems, covering traditional linear models, factorization machines, tree‑based GBDT+LR, and a range of deep learning architectures such as Wide&Deep, DeepFM, DCN, xDeepFM, DIN, as well as multi‑task frameworks like ESMM and MMOE, and finally illustrates their practical deployment in a live streaming platform.
This article explains the evolution of hotel aggregation at Mafengwo, from simple cosine similarity matching to advanced machine‑learning pipelines using tokenization, feature engineering, and LightGBM models, highlighting challenges of accuracy and real‑time performance and presenting practical solutions.
This article describes a complete pipeline for recommending automated test scripts for APK releases, covering CSV data preprocessing, feature encoding, tokenization with pkuseg and jieba, and training various machine‑learning models such as LDA, word2vec, XGBoost, deep neural networks, and multi‑label classifiers to predict script execution order.
This article presents a regression‑based approach that leverages historical bug‑fix data, feature selection, data cleaning, and model tuning to reorder redline scan warnings so developers can focus on the most critical fixes, improving overall repair efficiency.
This article details how Ctrip's search team leverages XGBoost for learning-to-rank, covering L2R concepts, feature engineering, data preparation, model training, hyper‑parameter tuning, offline evaluation, and deployment insights for large‑scale search ranking systems.
This article explains the importance of operations maintenance, outlines the challenges of traditional rule‑based anomaly detection, and describes how machine‑learning‑driven AIOps—including feature engineering, unsupervised and supervised models—can provide more accurate, scalable, and automated detection of server anomalies.
Effective short‑video recommendation relies on meticulous feature engineering that transforms raw signals—numerical counts, categorical IDs, content and user embeddings, context and session data—through bucketization, scaling, crossing, and smoothing, then selects and evaluates them via filtering, wrapping, regularization, and importance analysis to mitigate business biases and improve multi‑objective ranking performance.
