Semanlink - Information retrieval: techniques

Parents:

Information retrieval

Information retrieval: techniques

Related Tags:

Coursera: Web Intelligence and Big Data

Descendants

Embeddings in Information Retrieval
- DSSM (Deep Semantic Similarity Model)
Probabilistic relevance model
- BM25
- TF-IDF
Query Expansion
Ranking (information retrieval)
- BM25
- Discounted cumulative gain
- Google ranking
  - PageRank
- Learning to rank
  - Ranking SVM
  - TextRank
- Probabilistic relevance model
- Ranked Entities in Search Results
- Reranking
- Unsupervised ranker
Retriever
- Dense retriever
- Retriever-Reader
  - BLINK
  - RAG (Retrieval-Augmented Generation)
    - LLM Fine-tuning vs RAG
    - RAG: Agents
    - RAG: chunking
    - RAG + KG
    - RAG evaluation
    - RAG isn't easy
    - WalkingRAG
  - Retriever-Ranker
- Sparse retrieval
  - SPLADE
Shingles
Similarity queries / Vector search
- Approximate nearest-neighbor
  - faiss
  - HNSW
- Cosine similarity
- DSSM (Deep Semantic Similarity Model)
- Locality Sensitive Hashing
- Nearest neighbor search
  - Approximate nearest-neighbor
  - ElasticSearch: nearest neighbor(s)
  - faiss
  - HNSW
  - k-nearest neighbors
  - KNN in Multi-label classification
  - kNN-Transformers
- Probabilistic relevance model
- Semantic Hashing
TF-IDF
Vector Retrieval
- Similarity queries / Vector search
  [2401.09350] Foundations of Vector Retrieval
  Tags:
  2024-01-18 About
Vector space model
- gensim
- Latent Semantic Analysis
- TF-IDF
  [1910.03524] Beyond Vector Spaces: Compact Data Representation as Differentiable Weighted Graphs
  Tags:
  > In this paper, we aim to eliminate the inductive bias imposed by the embedding space geometry. Namely, we propose to map data into more general non-vector metric spaces: a weighted graph with a shortest path distance. By design, such graphs can model arbitrary geometry with a proper configuration of edges and weights. Our main contribution is PRODIGE (Probabilistic Differentiable Graph Embeddings): a method that learns a weighted graph representation of data end-to-end by gradient descent. [Github](https://github.com/stanis-morozov/prodige)
  
  2019-10-09 About
  [1706.00957] Semantic Vector Encoding and Similarity Search Using Fulltext Search Engines
  Tags:
  > The ubiquity of semantic vector space modeling raises the challenge of efficient searching in dense, high-dimensional vector spaces. We would naturally want to take advantage of the design and optimizations behind modern fulltext engines like Elasticsearch so as to meet the scalability and robustness demands of modern IR applications. This is the research challenge addressed in this paper. > The paper describes novel ways of encoding dense vectors into text documents, allowing the use of traditional inverted index engines. [blog post](https://rare-technologies.com/semantic-search-fulltext-engine-acl-2017/)
  
  2017-11-11 About
  From Frequency to Meaning: Vector Space Models of Semantics (2010)
  Tags:
  Good survey of VSMs, of their 3 classes (based either on term-document, word-context, or pair-pattern matrices), and of their applications. A detailed look at a specific open source project in each category.
  
  2017-07-10 About

2 Documents (Long List)

Fuzzy-Fingerprints for Text-Based Information Retrieval

Tags:

2013-05-31 About

Introduction to Information Retrieval, Cambridge University Press (2008) Manning, Raghavan, and Schütze

Tags:

2012-04-01 About

Properties

sl:creationDate : 2013-05-31
sl:creationTime : 2013-05-31T14:40:09Z
rdf:type : sl:Tag
skos:prefLabel : Information retrieval: techniques