Search Engines for Web and Enterprise Data

This module introduces the major tasks involved in web search, the history, evolution, impacts and challenges of web search engines.

This module covers different business models of web search engines, including keyword advertising, charging methods, and the business history of search engines.

This module explores different information retrieval models, including Boolean Models and Statistical models, and how to determine important words in a document using TFxIDF.

This module covers how to represent documents and queries as vectors of keywords, and how to determine the similarity between vectors using various measures like Inner Product, Cosine Similarity, Jaccard Coefficient, and Dice Coefficient.

This module explains how to index documents using inverted files, and how to perform update and deletion operations on inverted files. It also covers scalability issues and possible solutions.

This module covers how to use the Extended Boolean Model to rank documents and how to evaluate conjunctive and disjunctive queries using this model.

This module explores the history and evolution of link-based ranking methods, how to determine query/document similarities using HyPursuit, WISE, and PageRank, and possible extensions to PageRank.

This module explains how to calculate hub and authority scores of web documents using the HITS algorithm and the re-ranking process involved in HITS.

This module covers how to evaluate retrieval effectiveness using various measures like Precision, Recall, F-Measure, Average-Precision, DCG, and NDCG, as well as subjective relevance measures.

This module explains how to use the TREC collection for benchmarking and the characteristics of the TREC collection.

This module covers stemming, different Content-Sensitive and Context-Free stemming algorithms, and how to calculate Successor Variety and Entropy for stemming.

This module explains how to perform document space modification and query modification using relevance feedback.

This module covers the importance of relative preference in personalization, eye-tracking user studies, and modeling preferences as weighted vectors.

This module explains how to calculate discrimination value for index term selection and the importance of word usage in documents for search engine design.

This module covers using collocated terms instead of strict phrases in search, identifying collocated terms using Pointwise Mutual Information (PMI), and utilizing N-grams for search.

This module explores the challenges of enterprise search and the differences between web search and enterprise search.