Explicit semantic analysis (ESA) is a vectorial representation of text that uses a document corpus as a knowledge base.

In ESA, a word is represented as a column vector in the tf-idf matrix of the text corpus and a document is represented as the centroid of the vectors representing its words. Typicall, the text corpus is Wikipedia, though other corpora have been used.

1. Represent texts as weighted mix of predetermined set of natural concepts

2. Natrual concepts: Defined by humans, easily explained

Recommender System

1. Recommender System Overview
2. Collaborative filtering
   • Memory-based
     1. User-based
     2. Item-based
   • Model-based
3. Content-based filtering

Information Extraction

1. Approaches
   • Manually crafted patterns
   • Machine learning Approaches
2. Tasks and subtasks
3. Evaluation

Personalised Search

1. Personalized Search
   • Query Expansion
   • Search results filtering
   • Personal information gathering
     1. Implicit
     2. Explicit
   • Disadvantages
2. Personalized Mobile Search Engine
   • Client-Server architecture
     1. Client
     2. Server
   • Evaluation
3. PageRank
   • Two hypothesis
     1. Quantitative
     2. Qualitative
   • Initial Stage
   • Update Stage

Basic concepts.

1. Information Retrieval
2. Information Agent
   • Percepts
   • Actions
   • Goals
   • Environment
   • Drawbacks
3. Search Engine
   • What search engines do?
   • Indexing
   • Popular Retrieval models
4. Vector Space Model
   • Cosine Similarity
5. Text representation
   • Term weights
6. Evaluation
   • Criteria
   • How to evaluate two IR systems?