frutik/awesome-search: Awesome Search - this is all about the (e-commerce, but not only) search and its awesomeness

Awesome Search

Support Ukrainian fight for the freedom

I’ve been building e-commerce search applications for more than ten years. Below is a list of some publications, conferences, and books that have inspired me, grouped by topic. If an item fits into multiple topics, it appears in multiple sections.

⭐ Star us on GitHub — it helps!

Also check my other collections awesome e-commerce, awesome knowledge graphs, awesome cloud apps

Topics

General, fun, philosophy

• Falsehoods Programmers Believe About Search
• Ethical Search: Designing an irresistible journey with a positive impact
• On Semantic Search
• Feedback debt: what the segway teaches search teams
• Supporting the Searcher’s Journey: When and How
• Shopping is Hard, Let’s go Searching!
• An Introduction to Search Quality
• On-Site Search Design Patterns for E-Commerce: Schema Structure, Data Driven Ranking & More
• In Search of Recall
• Balance Your Search Budget!

Types of search

• Evolution of Search Technology: A Look Ahead

Classic/Lexical Search

• Etsy. Targeting Broad Queries in Search
• How Etsy Uses Thermodynamics to Help You Search for “Geeky”
• Broad and Ambiguous Search Queries
• Deconstructing E-Commerce Search: The 12 Query Types

Vectors/Semantic search

• Migrating to Elasticsearch with dense vector for Carousell Spotlight search engine
• From zero to semantic search embedding model
• Innovating Search Experience with Amazon OpenSearch and Amazon Bedrock

Symmetric and Asymmetric semantic search

• Symmetric vs. Asymmetric Semantic Search

Embeddings

Encoder architecture

Bi-encoders / Two towers (no interaction)

• Bi-encoder vs Cross encoder? When to use which one?

Cross-encoders (early interaction)

ColBERT (late interaction)

• Announcing the Vespa ColBERT embedder
• What is ColBERT and Late Interaction and Why They Matter in Search?

Vector types

Dense vectors

Input size limits

• Chunking Strategies for LLM Applications
• Evaluating the Ideal Chunk Size for a RAG System using LlamaIndex
• How to Chunk Text Data — A Comparative Analysis
  • Positional chunking
    • Semantic chunking

Matryoshka embeddings

• Matryoshka embeddings: faster OpenAI vector search using Adaptive Retrieval
• Introduction to Matryoshka Embedding Models
• Matryoshka representations. A guide to faster semantic search

Context-aware embeddings

• Improve your RAG applications by moving to Task-aware Embeddings
• How Context-Aware Embeddings Are Transforming Enterprise Search

Sparse vectors

SPLADE

• Hybrid Search: SPLADE (Sparse Encoder)
• SPLADE for Sparse Vector Search Explained
• Improving information retrieval in the Elastic Stack. Introducing Elastic Learned Sparse Encoder, our new retrieval model
• SPLADE – a sparse bi-encoder BERT-based model achieves effective and efficient first-stage ranking

Constructed query vectors

Hypothetical Document Embeddings (HyDE)

• HyDE

Bag-of-documents

• Distilling Retrieval Pipelines to a Single Embedding Model

Wormhole vectors

• Beyond Hybrid Search: Traversing Vector Spaces with Wormhole Vectors

Dimensionality handling

Dimensionality reduction

• PCA
• t-SNE

Quantization

• Scalar quantization
• Binary quantization
• Product quantization
• Rotational quantization

Finetuning

Supervised finetuning

• Fine-Tuning Text Embeddings For Domain-Specific Search
• Is Fine-Tuning an Embedding Model Worth it?

Knowledge distillation

• Distilling Retrieval Pipelines to a Single Embedding Model

Multimodal finetuning

• Fine-tuning Multimodal Embedding Models

Vector retrieval

• Choosing the best model for semantic search
• Guidelines to choose an index
• Nearest Neighbor Indexes for Similarity Search
• The Missing WHERE Clause in Vector Search

Hybrid search

• Hybrid search > sum of its parts?
• On Hybrid Search
• Hybrid search with Re-ranking

Reciprocal rank fusion (RRF)

• Hybrid search with Re-ranking
• Reciprocal rank fusion
• RRF is Not Enough

Linear Score Combination

• Linear retriever

Multimodal search

• Muves: Multimodal & multilingual vector search w/ Hardware Acceleration
• Model Selection for Multimodal Search

Multimodality Problems

Modality Gap

• Accept the modality gap: An exploration in the hyperbolic space

Contrastive Gap

• Characterizing and Addressing the Contrastive Gap

Agentic search

• Agentic Search as an Agile Engineering Process
• Agents turn simple keyword search into compelling search experiences
• Agentic search models

Search Quality Assurance

Evaluation Paradigms

Session-based Evaluation

• Session vs Query based search evals
• Measuring Search Effectiveness

Query-based Evaluation

Random sampling

• Simple random sample

Stratified sampling

• Stratified sampling
• The 8 Most Common Search Query Types

Probability-proportional-to-size sampling

• How to succeed with explicit relevance evaluation using Probability-Proportional-to-Size sampling

Metrics

• Choosing your search relevance evaluation metric
• Visualizing search metrics
• Choosing your search relevance evaluation metric
• Measuring Search: Metrics Matter

Focused on ranking quality

• Discounted cumulative gain
• Flavors of NDCG - normalized to what!?
• Mean reciprocal rank
• P@k
• Demystifying nDCG and ERR
• https://en.wikipedia.org/wiki/Precision_and_recall
• https://en.wikipedia.org/wiki/F1_score

Focused on diversity of results

• Diving into Diversity Metrics: Elevate Your Recommender Systems in a Snap!

MMR

• How to Calculate MMR?
• Maximal Marginal Relevance to Re-rank results in Unsupervised KeyPhrase Extraction

Average Pairwise Distance, APD

• edit distance
• semantic distance

Entropy

• How to measure Diversity of Search Results

Behavioral / Product / Performance

Clicks

Zero clicks

Clicks residual

• Click Residual: A Query Success Metric

Zero results

Evaluation Modes

Offline

• How to Implement a Normalized Discounted Cumulative Gain (NDCG) Ranking Quality Scorer in Quepid
• Compute Mean Reciprocal Rank (MRR) using Pandas

Judgements

HUman judgements

• What Is a Judgment List?
• Evaluating Search: Using Human Judgments
• Measuring Search, A Human Approach

Implicite judgements

add something on clicks streams

Using LLM as judge

• Improving retrieval with LLM-as-a-judge
• LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods

Online

• How we Compute NDCG in Daraz. E-Commerce

Areas of application

Enterprise search

• GenAI Can Improve Enterprise Search, But Remains a Work In Progress

e-Commerce search

• The influence of TF-IDF algorithms in eCommerce search

Conversational search

• Search as a Conversation
• Affordances for Conversational Search
• Query Understanding and Chatbots

Search Results

Retrieval

Relevance

• Humans Search for Things not for Strings
• What is a ‘Relevant’ Search Result?
• How to Achieve Ecommerce Search Relevance
• Setting up a relevance evaluation program

Relevance Algorithms

BM25

• Understanding the BM25 full text search algorithm
• Practical BM25: How Shards Affect Relevance Scoring in Elasticsearch, The BM25 Algorithm and its Variables
• The influence of TF-IDF algorithms in eCommerce search
• BM25 The Next Generation of Lucene Relevance
• Lucene Similarities (BM25, DFR, DFI, IB, LM) Explained

Bayesian BM25 (BB25)

• Bayesian BM25 is cool
• Releasing bb25 0.2.0: Why Bayesian BM25 (bb25) extends well far-beyond search?

Ranking

Multi-stage ranking

• Multi stage ranking
• Cross-Encoders, ColBERT, and LLM-Based Re-Rankers

Reranking

• Drowning in Documents: Consequences of Scaling Reranker Inference

Learning to Rank

• How is search different than other machine learning problems?
• Reinforcement learning assisted search ranking
• E-commerce Search Re-Ranking as a Reinforcement Learning Problem
• When to use a machine learned vs. score-based search ranker
• What is Learning To Rank?
• Using AI and Machine Learning to Overcome Position Bias within Adobe Stock Search
• Train and Test Sets Split for Evaluating Learning To Rank Models
• How LambdaMART works - optimizing product ranking goals

Click models for search

• Click models
• Click Modeling for eCommerce
• Using Behavioral Data to Improve Search

Bias

• What is Presentation Bias in search?
• Dealing with Position Bias in Recommendations and Search

Diversification

• Search Result Diversification using Causal Language Models
• Learning to Diversify for E-commerce Search with Multi-Armed Bandit
• Search Quality for Discovery & Inspiration
• How to measure Diversity of Search Results
• Searching for Goldilocks
• Broad and Ambiguous Search Queries - Recognizing When Search Results Need Diversification
• Thoughts on Search Result Diversity

MMR

• How to Calculate MMR?
• Maximal Marginal Relevance to Re-rank results in Unsupervised KeyPhrase Extraction

Personalisation

• Patterns for Personalization in Recommendations and Search
• Daniel Tunkelang Personalization
• Airbnb - Real-time personalization in search
• 98 personal data points that facebook uses to target ads to you
• Architecture of real world recommendation systems
• Feature engineering for personalized search

Zero search results

• Strategies for using alternative queries to mitigate zero results and their application to online marketplaces
• Semantic Equivalence of e-Commerce Queries

Search UX

Baymard Institute

• Deconstructing E-Commerce Search: The 12 Query Types
• Autodirect or Guide Users to Matching Category
• 13 Design Patterns for Autocomplete Suggestions (27% Get it Wrong)
• E-Commerce Search Needs to Support Users’ Non-Product Search Queries (15% Don’t)
• Search UX: 6 Essential Elements for ‘No Results’ Pages
• Product Thumbnails Should Dynamically Update to Match the Variation Searched For (54% Don’t)
• Faceted Sorting - A New Method for Sorting Search Results
• The Current State of E-Commerce Search
• E-Commerce Sites Need Multiple of These 5 ‘Search Scope’ Features
• E-Commerce Search Field Design and Its Implications
• E-Commerce Sites Should Include Contextual Search Snippets (96% Get it Wrong)
• E-Commerce Search Usability: Report & Benchmark
• Six ‘COVID-19’ Related E-Commerce UX Improvements to Make

Nielsen Norman Group

• The Love-at-First-Sight Gaze Pattern on Search-Results Pages
• Good Abandonment on Search Results Pages
• Complex Search-Results Pages Change Search Behavior: The Pinball Pattern
• Site Search Suggestions
• Search-Log Analysis: The Most Overlooked Opportunity in Web UX Research
• Scoped Search: Dangerous, but Sometimes Useful
• 3 Guidelines for Search Engine “No Results” Pages

Enterprise Knowledge LLC

• Optimizing Your Search Experience: A Human-Centered Approach to Search Design

Facets

• Facets of Faceted Search
• Coffee, Coffee, Coffee!
• Faceted Search (start here!)
• How to implement faceted search the right way
• Metadata and Faceted Search
• Metacrap: Putting the torch to seven straw-men of the meta-utopia
• 7 Filtering Implementations That Make Macy’s Best-in-Class
• Facet Search: The Most Comprehensive Guide. Best Practices, Design Patterns, Hidden Caveats, And Workarounds
• Facets: Constraints or Preferences?
• Facets, But Which Ones?

Accidental Taxonomist

• How Many Facets Should a Taxonomy Have
• When a Taxonomy Should not be Hierarchical
• Customizing Taxonomy Facets

Other

• Learning from Friction to Improve the Search Experience
• Why is it so hard to sort by price?
• Faceted Sorting
• Google kills Instant Search

Spelling correction

• Peter Norvig. “How to Write a Spelling Corrector”. Classic publication.
• Daniel Tunkelang. “Spelling Correction”
• A simple spell checker built from word vectors
• A closer look into the spell correction problem: 1, 2, 3, preDict
• Deep Spelling
• Modeling Spelling Correction for Search at Etsy
• Wolf Garbe. Author of Sympell. 1000x Faster Spelling Correction algorithm, Top highlight SymSpell vs. BK-tree: 100x faster fuzzy string search & spell checking, Fast Word Segmentation of Noisy Text
• Chars2vec: character-based language model for handling real world texts with spelling errors and
• JamSpell, spelling correction taking into account surrounding context - library, (in russian) Исправляем опечатки с учётом контекста
• Embedding for spelling correction
• A simple spell checker built from word vectors
• What are some algorithms of spelling correction that are used by search engines?
• Moman - lucene/solr/elasticsearch spell correction/autocorrect is (was?) actually powered by this library.
• Query Segmentation and Spelling Correction
• Applying Context Aware Spell Checking in Spark NLP
• Autocorrect in Google, Amazon and Pinterest and how to write your own one

Synonyms

• Boosting the power of Elasticsearch with synonyms
• Real Talk About Synonyms and Search
• Synonyms in Solr I — The good, the bad and the ugly
• Synonyms and Antonyms from WordNet
• Synonyms and Antonyms in Python
• Dive into WordNet with NLTK
• Creating Better Searches Through Automatic Synonym Detection
• Multiword synonyms in search using Querqy
• How to Build a Smart Synonyms Model
• The importance of Synonyms in eCommerce Search

Stopwords

• Do all-stopword queries matter?

Suggestions

Synonyms: autocomplete, search as you type, suggestions

• Giovanni Fernandez-Kincade. Bootstrapping Autosuggest, Building an Autosuggest Corpus, Part 1, Building an Autosuggest Corpus, Part 2, Autosuggest Retrieval Data Structures & Algorithms, Autosuggest Ranking
• On two types of suggestions
• Improving Search Suggestions for eCommerce
• Autocomplete Search Best Practices to Increase Conversions
• Why we’ve developed the searchhub smartSuggest module and why it might matter to you
• Nielsen Norman Group: Site Search Suggestions
• 13 Design Patterns for Autocomplete Suggestions
• Autocomplete
• Autocomplete and User Experience
• IMPLEMENTING A LINKEDIN LIKE SEARCH AS YOU TYPE WITH ELASTICSEARCH
• Smart autocomplete best practices: improve search relevance and sales
• OLX: Building Corpus for AutoSuggest (Part 1), AutoSuggest Retrieval & Ranking (Part 2)
• Autocomplete, Live Search Suggestions, and Autocorrection: Best Practice Design Patterns
• Mirror, Mirror, What Am I Typing Next? All About Search Suggestions
• How we built the lightning fast autosuggest for otto.de

Graphs/Taxonomies/Knowledge Graph

• Knowledge graphs applied in the retail industry Knowledge graphs are becoming increasingly popular in tech. We explore how they can be used in the retail industry to enrich data, widen search results and add value to a retail company.
• Awesome Knowledge Graphs

Integrating Search and Knowledge Graphs (by Enterprise Knowledge)

• Part 1: Displaying Relationships
• Search query expansion with query embeddings

Query expansion

• Fundamentals of query rewriting (part 1): introduction to query expansion

Query understanding

• Daniel Tunkelang Query Understanding.
• Query Understanding, Divided into Three Parts
• Search for Things not for Strings
• Understanding the Search Query. Part 1, Part 2, Part 3
• Food Discovery with Uber Eats: Building a Query Understanding Engine
• AI for Query Understanding

Search Intent

• Mapping Search Queries To Search Intents
• Search: Intent, Not Inventory

Query segmentation

• Paper Unsupervised Query Segmentation Using only Query Logs
• Paper Towards Semantic Query Segmentation

Algorithms

BERT

• Understanding BERT and Search Relevance
• Google is improving web search with BERT – can we use it for enterprise search too?

ColBERT

• Pretrained Transformer Language Models for Search - part 3

Collocations, common phrases

• Automatically detect common phrases – multi-word expressions / word n-grams – from a stream of sentences.
• The Unreasonable Effectiveness of Collocations

Other Algorithms

• One hot encoding
• Writing a full-text search engine using Bloom filters

Hashing

• Locality Sensitive Hashing
• Locality Sensitive Hashing (LSH): The Practical and Illustrated Guide
• Minhash

Sorting by average ratings

• Better than Average: Sort by Best Rating
• How Not To Sort By Average Rating

Keywords extraction

• Keyword Extraction using RAKE
• Yet Another Keyword Extractor (Yake)
• Keyword Extraction with BERT

Tracking, profiling, GDPR, Analysis

Tools, platforms, helpers for search tracking

• OpenSearch User Behavior Insights
• Site Search tracking with Google Analytics 4
• Snowplow
• search-colletor
• OpenTelemetry with search additions
• Pulse Query Analytics
• Tracking who’s hot and who’s not presents an algorithmic challenge

Resources

• Anonymisation: managing data protection risk (code of practice)
• The Anonymisation Decision-Making Framework
• 98 personal data points that facebook uses to target ads to you
• Opportunity Analysis for Search
• A Face Is Exposed for AOL Searcher No. 4417749
• AOL search data leak
• Personal data

Experiments

• Common Pitfalls of Search Experimentation
• Improving Search @scale with efficient query experimentation

A/B testing, MABs

• A/B Testing for Search is Different
• A/B Testing Search: thinking like a scientist

Testing, metrics, KPIs

KPIs

• 5 Right Ways to Measure How Search Is Performing
• E-commerce Site-Search KPIs. Part 1 – Customers, Part 2 – Products, Part 3 - Queries
• Learning from Friction to Improve the Search Experience
• Behind the Wizardry of a Seamless Search Experience
• Analyzing online search relevance metrics with the Elastic Stack
• How to Gain Insight From Search Analytics

Evaluating Search (by Daniel Tunkelang)

• Measure It
• Measuring Searcher Behavior
• Using Human Judgement
• When There’s No Conversion Rate

Measuring Search (by James Rubinstein)

• Statistical and human-centered approaches to search engine improvement
• A Human Approach
• Setting up a relevance evaluation program
• Metrics Matter
• A/B Testing Search: thinking like a scientist
• Query Triage: The Secret Weapon for Search Relevance
• The Launch Review: bringing it all together…

Three Pillars of Search Relevancy (by Andreas Wagner)

• Part 1: Findability
• part 2: Search Quality For Discovery & Inspiration

Architecture

• The Art Of Abstraction – Revisiting Webshop Architecture
• Canva - Search Pipeline
  • Part One outline of the challenges faced
    • Part Two new search arcthitecture
• Event-Driven Architecture for Efficient Search Indexing

Education and networking

Events

• Search Conference Calendar - The Search Juggler

Conferences

• Activate
• Berlin buzzword
• Haystack
• Elastic{ON}
• MIX-CAMP E-COMMERCE SEARCH
• SIGIR eCommerce
  • 2019
    • 2018
    • 2017

Trainings and courses

• Cheat at Search with LLMs. Doug Turnbull Next: July 2025
• Relevant Search Masterclass by Doug Turnbull Next: July 2025
• OpenSource Connections
  • Elasticsearch “Think Like a Relevance Engineer”
    • Solr “Think Like a Relevance Engingeer”
    • Beyond Search Relevance: Understanding and Measuring Search Result Quality
    • Hello LTR
• Sease’s trainings
• Search Fundamentals. Daniel Tunkelang, Grant Ingersoll Next: Feb 6, 2023
• Search with Machine Learning. Daniel Tunkelang, Grant Ingersoll Next: Feb 27, 2023
• Search for Product Managers. Daniel Tunkelang Next: Apr 3, 2023
• Sematext’s Solr, Elasticsearch, and OpenSearch trainings Fall 2023
• https://dtunkelang.medium.com/upcoming-search-classes-this-fall-58f877fe00ad

Books

• AI-powered search
• Relevant Search
• Deep Learning for search
• Interactions with search systems
• Embeddings in Natural Language Processing. Theory and Advances in Vector Representation of Meaning
• Search User Interfaces
• Search Patterns
• Search Analytics for Your Site: Conversations with Your Customers
• Click Models for Web Search
• Optimization Algorithms
• Query Understanding for Search Engines

Blogs and Portals

• Searchnews

Papers

• List of papers

Search Team. Managment, composition, hiring

• Search is a Team Sport
• Thoughts about Managing Search Teams
• On Search Leadership
• Building an Effective Search Team: the key to great search & relevancy
• Query Triage: The Secret Weapon for Search Relevance
• The Launch Review: bringing it all together
• The Role of Search Product Owners
• Search Product Management: The Most Misunderstood Role in Search?
• Search relevance for understaffed teams

Job Interviews

• Interview Questions for Search Relevance Engineers, Data Scientists, and Product Managers
• Data Science Interviews: Ranking and search

Engineering

• Technical debt in search

Economics of Search

Blogposts series

Search Optimization 101 (by Charlie Hull)

• How do I know that my search is broken?
• What does it mean if my search is ‘broken’?
• How do you fix a broken search?
• Reducing business risk by optimizing search

Query Understanding (by Daniel Tunkelang)

Better search through query understanding.

• An Introduction
• Language Identification
• Character Filtering
• Tokenization
• Spelling Correction
• Stemming and Lemmatization
• Query Rewriting: An Overview
• Query Expansion
• Query Relaxation
• Query Segmentation
• Query Scoping
• Entity Recognition
• Taxonomies and Ontologies
• Autocomplete
• Autocomplete and User Experience
• Contextual Query Understanding: An Overview
• Session Context
• Location as Context
• Seasonality
• Personalization
• Search as a Conversation
• Clarification Dialogues
• Relevance Feedback
• Faceted Search
• Search Results Presentation
• Search Result Snippets
• Search Results Clustering
• Question Answering
• Query Understanding and Voice Interfaces
• Query Understanding and Chatbots

Grid Dynamics

• Not your father’s search engine: a brief history of retail search
• Semantic vector search: the new frontier in product discovery
• Boosting product discovery with semantic search
• Semantic query parsing blueprint

Considering Search: Search Topics (by Derek Sisson)

• Intro
• Assumptions About Search
• Assumptions About User Search Behavior
• Types of Information Collections
• A Structural Look at Search
• Users and the Task of Information Retrieval
• Testing Search
• Useful Search Links and References

Industry players

Personalies and influencers

• Daniel Tunkelang (he is God of Search)
• Max Irwin
• Doug Turnbull
• Baymard’s Institute

Search Engines

• Google
• Bing
• Not Human Search - Agent-first search engine for discovering AI tools and MCP servers
• Amazon
• eBay

Products and services

• Algolia
• [Vespa] (https://vespa.ai/)
• Elasticsearch - Distributed search & analytics engine
• ParadeDB - Modern Elasticsearch alternative built on Postgres. Built for real-time, update-heavy workloads.
• Solr - Solr is the blazing-fast, open source, multi-modal search platform built on the full-text vector, and geospatial search capabilities of Apache Lucene
• Fess Enterprise Search Server
• Typesense - an opensource alternative to Algolia.
• TopK - combines AI-powered query understanding with adaptive ranking to provide the most relevant results in your domain.
• SearchHub.io
• Datafari - an open source enterprise search solution.
• Qdrant - an open source vector database.
• Awakari - Real-Time search from unlimited sources like RSS, Fediverse, Telegram. Text keyword matching conditions, numeric conditions, condition groups. Reverse search index based.
• Meilisearch - Open source search API that supports full-text, vector, geospatial & faceted search.
• Omnigraph - Typed graph database where agents branch and merge like Git. S3-native, Rust, traversal + vector + BM25 in one runtime.
• SearchPixel - Hybrid CLIP + BGE product-search API for Shopify and WooCommerce stores. Combines visual and semantic search, priced in INR (Indian rupees), and free during the open beta for the first 50 stores.

Consulting companies

• BigData Boutique
• OpenSource Connections
• https://sease.io/
• Sematext

Case studies

• Airbnb - Machine Learning-Powered Search Ranking of Airbnb Experiences
• Airbnb - Listing Embeddings in Search Ranking
• Algolia - The Architecture Of Algolia’s Distributed Search Network
• Meituan - Exploration and practice of BERT in the core ranking of Meituan search (🇨🇳 BERT在美团搜索核心排序的探索和实践)
• Netflix - How Netflix Content Engineering makes a federated graph searchable (Part 1, Part 2)
• Netflix - Elasticsearch Indexing Strategy in Asset Management Platform (AMP)
• Skyscanner - Learning to Rank for Flight Itinerary Search
• Slack - Search at Slack
• Twitter - Stability and scalability for search
• Amazon SEO Explained: How to Rank Your Products #1 in Amazon Search Results in 2020
• Building a Better Search Engine for Semantic Scholar

General search

• How Bing Ranks Search Results: Core Algorithm & Blue Links
• How Google Search Ranking Works – Darwinism in Search

E-commerce

• Searchandising

Multisided markets

• Discover How Cassini (The eBay Search Engine) Works and Rank

Videos

Apache Solr Short Tips

Channels

• Lucid Thoughts
• Lucidworks
• MIx-Camp E-commerce Search
• OpenSource Connections
• SIGIR eCom

Featured

• Relevant Facets

Datasets

• Shopping Queries Dataset: A Large-Scale ESCI Benchmark for Improving Product Search
• ESCI-S: extended metadata for Amazon ESCI dataset
• Home Depot Product Search Relevance
• WANDS - Wayfair ANnotation Dataset

Tools

Spacy

Awesome Spacy - Natural language upderstanding, content enrichment etc.

Word2Vec

• Word2Vec For Phrases — Learning Embeddings For More Than One Word
• Gensim Word2Vec Tutorial
• How to incorporate phrases into Word2Vec – a text mining approach
• Word2Vec — a baby step in Deep Learning but a giant leap towards Natural Language Processing
• How to Develop Word Embeddings in Python with Gensim

Libs

• Query Segmenter
• Tantivy - Tantivy is a full-text search engine library inspired by Apache Lucene and written in Rust
• https://github.com/zentity-io/zentity
• https://github.com/mammothb/symspellpy
• https://github.com/searchhub/search-collector
• Kiri - State-of-the-art semantic search made easy.
• Haystack - End-to-end Python framework for building natural language search interfaces to data.
• https://github.com/castorini/docTTTTTquery
• https://github.com/AiDinho/ruosh - A full text search library with Tantivy backend,all the goodness of whoosh api powered by Tantivy

Other

• Chorus, Smui, Querqy
• DocKit - Desktop GUI client for Elasticsearch, OpenSearch, and DynamoDB with AI-powered query generation
• Quepid
• Rated Ranking Evaluator
• Jina AI - A neural search framework

Other awesome stuff

• Awesome Knowledge Graphs
• Awesome time series
• Awesome Spacy
• Query-Understanding
• Click models

Unsorted

• IndexFox - AI-powered site search SaaS combining keyword and semantic search with instant AI-generated answers.
• sandbox Jun 2021
• sandbox May 2021
• sandbox April 2021
• sandbox Dec 2020
• sandbox Jan 2020

OLD TOC (to review)

• General, fun, philosophy
• Types of search
  • Classic/Lexical search
    • Vectors/Semantic search
    • Embeddings
      • Types of vectors
        • Dense vectors
          • Size of input and Chunking
            • Positional chunking - Semantic chunking - Matryoshka embeddings - Sparse vectors
          • SPLADE
            • Vector retrieval
        • Main architectures - Query/Document tokens interaction
          • No interactions - Two towers / Bi-encoders - Early interactions - Cross-encoders - Late interactions - ColBERT
            • Handling high-dimension embeddings
        • Dimensionality reduction - Quantization
          • Scalar quantization - Binary quantization - Product quantization - Rotational quantization
            • Finetuning models
        • Symmetric and Asymmetric semantic search
    • Hybrid search
    • Reciprocal rank fusion - RRF - Linear Score Combination
    • Multimodal search
    • Multimodality Problems
      • Modality Gap - Contrastive Gap
    • Federated search
• Search Quality Assurance
  • How to select queries
    • Random sampling - Stratified sampling - Probability-proportional-to-size sampling
    • Metrics
    • Focused on ranking quality - Focused on diversity of results
    • Offline quality measuring
    • Judgements
      • Explicite judgements, HUman judgements - Implicite judgements - Using LLM as judge
    • Online quality measuring
• Areas of application
  • Enterprise search
    • e-Commerce search
    • Conversational search
    • Geo-Spatial Search
    • Medical and Healthcare Search
    • Social Media and User-Generated Content Search
    • Question Answering Systems
    • Personal Information Management
• Search Results
  • Retrieval
    • Relevance
      • Relevance Algorithms
    • Ranking
    • Multi stage ranking
      • Reranking
        • Learning to Rank
      • Click models for search
    • Bias
    • Diversification
    • MMR
    • Personalisation
    • Zero search results
• Search UX
  • Baymard Institute
    • Nielsen Norman Group
    • Enterprise Knowledge LLC
    • Facets
    • Accidental Taxonomist
    • Other
• Spelling correction
• Suggestions
• Synonyms
• Stopwords
• Graphs/Taxonomies/Knowledge Graph
  • Integrating Search and Knowledge Graphs (by Enterprise Knowledge)
• Query expansion
• Query understanding
  • Search Intent
    • Query segmentation
• Algorithms
  • BERT
    • ColBERT
    • Collocations, common phrases
    • Other Algorithms
• Tracking, profiling, GDPR, Analysis
• Experiments
  • A/B testing, MABs
• Evaluating search
  • MRR
    • Testing, metrics, KPIs
    • KPIs
    • Evaluating Search (by Daniel Tunkelang)
    • Measuring Search (by James Rubinstein)
    • Three Pillars of Search Relevancy (by Andreas Wagner)
• Architecture
• Vectors search
• Education and networking
• Management, Search Team
  • Job Interviews
    • Engineering
• Industry players
  • Personalies and influencers
    • Search Engines
    • Products and services
    • Consulting companies
• Blogposts series
  • Search Optimization 101 (by Charlie Hull)
    • Query Understanding (by Daniel Tunkelang)
    • Grid Dynamics
    • Considering Search: Search Topics (by Derek Sisson)
• Videos
  • Channels
    • Featured
• Case studies
• Datasets
• Tools