Over the course of three posts I am going to round-up the various patent awards (of interest) from the 2010. This is a habit I've had since 2008 and it is a nice way for me to track things for future reference. While it's not exactly entertaining, I hope you bookmark and make use of them in the future.

Today we're getting things started with Yahoo's offerings. Yes, their search has become somewhat redundant these days, but search is search and there's always tidbits of interest. These aren't all of the patents awarded, but the ones I felt were relevant to the search geeks of the world.

The largest area was in the query analysis sector and this is fairly standard for Yahoo as they have some of the best computer scientists in this area. This is followed up by semantic analysis and local which are also quite prevalent among ALL of the engines. If you didn't already guess that local is huge, this adds more fuel to that fire.

Anyway, on with the list...

Query Analysis

• Using activation paths to cluster proximity query results
• Predictions of a degree of relevance between query re-writes and a search query
• Identifying related searches in a database search system
• Query categorizer
• Speculative search result based on a not-yet-submitted search query
• System and method for query expansion
• Techniques for navigational query identification
• Query log mining for detecting spam hosts
• Query Aware Processing
• Query difficulty estimation
• Method and apparatus for reconstructing a search query
• System and method for generating substitutable queries on the basis of one or more features
• Context-sensitive query expansion
• Search systems and methods using enhanced contextual queries
  
  

Semantic and NLP

• Background contextual conversational search
• Method and apparatus using a classifier to determine semantically relevant terms
• Cross-lingual query classification
• Phonetic self-improving search engine
• Searching document collections using semantic roles of keywords
• System for classifying words
• Determining Semantically Related Terms Using an Active Learning Framework
• Discovering alternative spellings through co-occurrence
• System and method for determining semantically related terms based on sequences of search queries
• Predictive stemming for web search with statistical machine translation models

Ranking methods

• Adapting a ranking function of a search engine for use with a specific domain
• Pre-fetching data for document ranking
• Conjoint Analysis with Bilinear Regression Models for Segmented Predictive Content Ranking
• Learning a ranking model that optimizes a ranking evaluation metric for ranking search results of a search query
• Ranking documents
• Using specificity measures to rank documents
• Reconstructing a search query
• Categorizing or selecting potential search results
• Predicting and ranking search query results
• Learning ranking functions incorporating isotonic regression for information retrieval and ranking

Geo Local

• Search results for local versus traveller
• System and method for associating a geographic location with an internet protocol address
• System for determining the geographic range of local intent in a search query
• Method of determining geographical location from IP address information
• Local query identification and normalization for web search
• System for providing geographically relevant content to a search query with local intent
• Entity based search results and clusters on maps
• System for determining local intent in a search query
• Geographical location extraction
  
  

Behavioural data

• System and method for indexing web content using click-through features
• User Sensitive PageRank
• Classifying documents using implicit feedback and query patterns
• Building a research document based on implicit/explicit actions
• Prediction of click through rates using hybrid kalman filter-tree structured markov model classifiers
• Personalized search results to multiple users
• Search Results for a topic that include a list of messaging users that have previously searched that topic.
• System and Method for Recommending Personalized Identifiers
• System and method for modeling user selection feedback in a search result page
  

Links

• Link optimization
• Determining related terms based on link annotations of documents belonging to search result sets
• Method And System For Generating A Hyperlink-Click Graph
• Enriched document representation using aggregated anchor text
  
  

Social

• Federated community search
• System and method for ranking search results using social information
• Conducting a profile based search

Recommendation engine

• Assisting search requests with vertical suggestions
• Suggesting an alternative to the spelling of a search query
• System and method for identifying related queries for languages with multiple writing systems
• System and method for generating substitutable queries on the basis of one or more features
• Search query concept based recommendations
  
  

Page Segmentation

• Techniques for clustering structurally similar web pages based on page features
• Techniques for clustering structurally similar web pages
• Method for improving quality of search results by avoiding indexing sections of pages
  
  

Universal

• Universal search interface systems and methods
• Blending Search Results
• Vertical search expansion, disambiguation, and optimization of search queries
• Approach for near duplicate image detection
  
  

Semantic mark-up

• Use of extensible markup language in a system and method for influencing a position on a search result list generated by a computer network search engine
• Semantic Search Extensions for web search engines
  
  

Mobile

• Suggesting keywords based on mobile specific attributes
• Enabling clustered search processing via text messaging
  

Duplicate content

Techniques for detecting duplicate web pages

Web spam

Detection of undesirable web pages

Temporal

Identifying and expanding implicitly temporally qualified queries

Systemic

• System and method for navigating and indexing content
• Interactive system for internet information retrieval and exploration
• Information retrieval system and method employing spatially selective features
• URL based query for retrieving data related to a context
• Method and system for quantifying the quality of search results based on cohesion
• Information retrieval engine
• Navigation assistance for search engines
• Predictive indexing for fast search
• Configuring a search engine results page with environment-specific information
• Query based search engine
• Method and system for web searching
• System and method for knowledge based search system
• System and method for generating expertise based search results
• Systems and methods for indexing content for fast and scalable retrieval

Other

• Generation of search result abstracts
• Selectively performing searches against data and providing search results
• Rare query expansion by web feature matching
• Web page categorization using graph-based term selection
• Methods of processing and segmenting web usage information
• Selecting and presenting user search results based on an environment taxonomy

And there we have it. These lists are very handy when you are researching a given aspect of search for your programs. This is how I use them actually... collect methods from the main engines and analyze the approaches to get a feel for how they are doing things.

But, please do remember; patents are the past, papers are the future. Use these for insight but remember some don't get used, some are many years old and they have likely evolved over time.

Next time out we will be looking at the offerings from Microsoft and then we will get to the big dog that is Google...

More articles by this author

Google Hacks & Dorks for fun and profit

Recently someone was asking me about Google's advanced operators and...

Read More >>

Latest Google Algorithm change

It's all about attribution On Friday it seems (Google's) Matt...

Read More >>