Please enable JavaScript.
Coggle requires JavaScript to display documents.
Query Representation - Coggle Diagram
Query Representation
Queries as Vectors
Concept of Vector Space Model (VSM)
Representation of documents and queries in n-dimensional space
Term weighting in vectors
Cosine similarity for comparing vectors
Computing Vector Scores
Dot product calculation between document and query vectors
Normalization of vectors
Handling missing terms or zero-weight terms
Use of sparse matrix representations
Efficient Scoring & Ranking
Use of inverted index for scoring
Heap-based sorting for top results
Skipping and impact-based pruning
Score caching and early termination techniques
Inexact Top-K Document Retrieval
Need for efficiency over exact ranking
Threshold algorithms (e.g., Fagin’s Algorithm)
Trade-off between accuracy and performance
Use in large-scale search engines