Facing rapid business growth and limited after‑sales resources, a company built an RPA‑driven intelligent WeChat customer service that monitors hundreds of groups, dynamically routes queries to a knowledge base or human agents, and operates 24/7 to improve response speed and reduce workload.
Customer Service AutomationEnterprise MessagingRPA
This article presents Nine Chapters Cloud Extreme DataCanvas's insights and practices on multimodal large model platforms, covering their historical development, platform components such as AI Foundation Software and Prompt Manager, practical implementations like memory-augmented models and ETL pipelines, and future prospects for enterprise knowledge bases and agents.
This article describes the design and implementation of a code‑review assistant powered by open‑source large language models and a privately hosted knowledge base, covering background, pain points, system architecture, model selection, vector‑store integration, prompt engineering, diff parsing, and practical reflections.
This article explores how combining traditional keyword search with large language models (LLMs) enhances understanding of user intent, builds a robust automotive knowledge base, and delivers more accurate, context‑aware answers through a multi‑stage retrieval and generation pipeline.
This article explores how vector databases serve as the memory layer for large AI models, detailing their distributed, compute‑separated architecture, performance optimizations, hybrid vector‑scalar retrieval, and practical deployments across TikTok’s ecosystem such as image search, intelligent Q&A, and multimodal AI services.
This article walks through constructing a domain‑specific, soul‑healing chatbot using LangChain and Llama 2, comparing fine‑tuning versus external knowledge bases, detailing environment setup, data loading, text splitting, embedding with a Chinese model, vector store creation, prompt engineering, inference, and optimization strategies.
The author builds an internal code knowledge base for the TDesign Vue‑Next library by scraping documentation, chunking and embedding texts with OpenAI’s ada model into a vector store, then retrieving relevant chunks for LLM answers, and enhances context continuity using a JavaScript AST interpreter, achieving up to 90 % query accuracy and a 20 % productivity boost.
This article explains what embeddings are, how they capture semantic similarity, and how to use OpenAI's embedding API to transform a custom knowledge base into vector representations for efficient search, retrieval, and question‑answering.
This study presents a comprehensive evaluation of ChatGPT's ability to answer complex knowledge‑base questions by introducing a feature‑driven multi‑label classification framework and applying the CheckList black‑box testing methodology across eight KB‑based CQA datasets, comparing its performance with GPT‑3, GPT‑3.5 and FLAN‑T5.
The article explains AIOps concepts, details China Mobile Information's award‑winning intelligent operations tools, shares an interview with deputy general manager Liang Enlei on their development, evaluation experience, future plans, and introduces the national AIOps maturity model and its key modules.
The interview details Yuque's inception as an internal Ant Group project, its evolution from a document tool to a structured knowledge‑base platform, its differentiation from Google Docs and GitHub, internal adoption within Alibaba, relationships with DingTalk and Alipay, the concept of a digital garden, and its product positioning for individuals and small‑to‑medium enterprises.
This article examines the technical challenges of traditional e‑commerce live streaming, presents a 24‑hour AI‑driven virtual live‑broadcast system built by Yanxuan and NetEase Fuxi, and details its architecture, virtual‑human generation, automated content creation, intelligent interaction, and future expansion plans.
This article examines the shortcomings of traditional regression testing, introduces a precise testing framework that combines black‑box and white‑box techniques, outlines its core principles, evolution steps, and measurement methods, and shares practical outcomes and future directions for server‑side quality assurance.
This guide walks you through using Yuque, a cloud‑based knowledge base, to quickly create, customize, and publish open‑source project documentation without coding or server setup, covering core concepts, step‑by‑step building, public access configuration, and optional homepage customization.
This article summarizes a conference presentation on business continuity management, covering its definition, pandemic‑driven opportunities, core concepts, best‑practice framework, monitoring, change control, knowledge‑base integration, and real‑world case studies, emphasizing its critical role for modern cloud‑based operations.
The ZhiXiaoBao technical team at Ant Financial broke the SQuAD 2.0 leaderboard with a model that surpasses human performance, detailing the challenges of natural‑language understanding, the specific ranking and data‑augmentation techniques they employed, and the broader impact on fintech knowledge‑base automation and future AI research.
This article explains how customer‑service chatbots use well‑structured knowledge bases and advanced sentence‑similarity algorithms to achieve over 90% question‑matching rates, covering knowledge‑base construction, maintenance, and the core AI models that power accurate responses.
The article explains how to construct an effective DBA operations team by focusing on reusable infrastructure, clear team standards, a structured training system, a comprehensive knowledge base, and a positive team atmosphere, providing practical tools and methods for each aspect.
This article presents the design, implementation, and performance of 58.com’s dialogue robot for local life services, detailing its business background, overall AI architecture, core QABot and TaskBot capabilities, intelligent lead generation, opportunity mining, and future research directions.
The article reviews the open‑source qa_match tool from 58.com, detailing its deep‑learning based question‑answer matching architecture, hierarchical knowledge‑base support, lightweight pre‑training model SPTM, and practical applications, while summarizing the live‑stream presentation and Q&A session.
The article introduces qa_match V1.1, an open‑source deep‑learning QA matching tool that adds one‑level knowledge‑base support, releases a lightweight Bi‑LSTM pre‑trained language model (SPTM), and provides detailed architecture, training data, performance benchmarks, future plans, and contribution guidelines.
The article introduces qa_match V1.1, an open‑source, Apache‑licensed lightweight question‑answer matching system that adds a simple pre‑trained language model (SPTM), supports one‑level knowledge bases, details model architecture, training resources, performance benchmarks, future plans, and contribution guidelines.
This article outlines an AI-powered workflow for constructing and enriching a business knowledge base in intelligent customer service, covering preprocessing, intent detection, deep and shallow semantic feature engineering, hierarchical bucket clustering, and automated summary extraction to improve FAQ coverage and reduce manual workload.
This article presents a comprehensive study of applying AI, big‑data analytics, and automated pipelines to improve operational efficiency in the securities industry, detailing a custom intelligent ops platform, its layered architecture, and three real‑world scenarios—root‑cause analysis, knowledge‑base assistance, and capacity forecasting—along with experimental results and practical insights.
The article details the design and iterative improvement of 58.com’s intelligent QABot, covering knowledge‑base construction, feature engineering, three generations of classification models—including FastText, Bi‑LSTM, and deep semantic matching—and evaluation metrics that achieve high accuracy and automation rates.
This article explains how Baidu Cloud's Noah intelligent operations product builds a unified operations knowledge base by categorizing metadata, status, and event data and applying three ETL approaches—Pull, Push, and Lazy—to handle offline, near‑line, and real‑time data integration.
The article details the design and implementation of Xiaomi's music‑focused natural language understanding platform, covering its service architecture, intent extraction, knowledge‑base search, slot filling, personalization, and the specific data and modeling challenges encountered.
The paper presents two inexpensive techniques for extracting geographic aliases of points of interest—comparing high‑dimensional contextual vectors of nearby shipping addresses and analyzing co‑occurring words in identical addresses—to construct a knowledge base that links official names with their synonyms, improving location‑based service accuracy.
This article summarizes the award‑winning CIKM 2017 paper “Hike: A Hybrid Human‑Machine Method for Entity Alignment in Large‑Scale Knowledge Bases,” explaining its motivation, the entity‑partition and predicate‑similarity techniques, the construction of partial orders, question‑selection strategies, inference modeling, error‑tolerance mechanisms, and the greedy algorithms that together achieve state‑of‑the‑art alignment performance.
The article details Qunar's development of an AI-powered customer service robot, describing its motivation, data analysis, multi‑phase cognitive system architecture, knowledge‑base management, evaluation mechanisms, and future integration into a group‑wide intelligent service platform to improve service efficiency and reduce costs.
This article explains how building complete, accurate, and real‑time foundational data—covering CMDB, logs, databases, and knowledge bases—forms the essential base for reliable IT operations, detailing the three core attributes, practical collection methods, and the structure of each data module.
In a June 28 presentation at Baidu’s 51st Technology Salon, senior NLP researcher Zhao Shiqi showcased Baidu Kaoba—a conversational search tool for college admissions—illustrating how inference‑based rewriting, ellipsis completion, concept clarification and interactive generation enable deep, domain‑specific semantic understanding that he predicts will power future internet services.