Next up in our patent round ups for 2010 is the only other big dog left standing; Microsoft. The laywers over there must keep REAL busy as there are far more than just search patents being pumped out each month (think computing, mobile, peripherals, Xbox and more). Actually makes collecting these harder than it’s counterparts at Google/Yahoo.
Given that MS is now handling Yahoo’s search, understanding how they do things has doubled in importance over the last 12 months. Once more semantic and NLP related patents are the largest sector with MS. This is actually par for the course over there. It, like query analysis at Yahoo, is an area they’ve always been strong in. There’s also a strong showing from the ‘Ranking Methods’ and ‘Behavioural Data’ areas. Behavioural has been big with all the contenders over the last few years.
On the lite-side of things, there was a distinct lack of geo-local stuff which seems odd given the movement in those areas over the last while. But hey, these are just awards right? We can’t read TOO much into it. I also found the presence of ‘named entities’ and a lack of link related offerings, to be potentially telling as well.
Anyway, here’s this year’s list for those keeping score;
Semantic and NLP
- Keyword usage score based on frequency impulse and frequency weight
- Fast on-line learning of transformed hidden Markov models
- Document summarization by maximizing informative content words
- System and method for context sensitive searching
- Execution of semantic queries using rule expansion
- Modeling semantic and structure of threaded discussions
- Text analysis using phrase definitions and containers
- Learning Latent Semantic Space for Ranking
- Synonym and similar word page search
- Identifying conceptually related terms in search query results
- Synonym and similar word page search
- Lexicon-based content correlation and navigation
Ranking methods
- Rank document based on a popularity of key meta-data
- UserRank: ranking linked nodes leveraging user logs
- Ranking results using multiple nested ranking
- Ranking search results using feature extraction
- System and method for incorporating anchor text into ranking search results
- Systems and methods for ranking documents based upon structurally interrelated information
- Ranking results based on query specific position bias
- Learning Latent Semantic Space for Ranking
- Systems and methods for ranking documents based upon structurally interrelated information
- Learning a document ranking function using fidelity-based error measurements
- Web page ranking with hierarchical considerations
- Training a ranking component
- Re-ranking top search results
- Modifying relevance ranking of search result items
- Ranking System
- Performing searches and returning results based on weighted criteria
Behavioural
- Recommendation system that identifies a valuable user action by mining data supplied by a plurality of users to find a correlation that suggests one or more actions for notification
- Adapting search results to personal information needs
- Calculating web page importance based on web behaviour model
- Mining translations of web queries from web click-Through data
- Accounting for behavioral variability in web search
- Ranking documents with social tags
- Augmenting user, query, and document triplets using singular value decomposition
- Recommendation of related electronic assets based on user search behavior
- Personalized information retrieval search with backoff
- Smooting click through data for web search ranking
- User-defined relevance ranking for search
Links
- Search Engine Enhancement using mined implicit links
- Pseudo-anchor text extraction for vertical search
- Visualizing hyperlinks in a search result list
- System and method for incorporating anchor text into ranking search results
- Using link structure for suggesting related queries
Query Analysis
- Searching questions based on topic and focus
- Predicting future queries from log data
- Relevant navigation with deep links into query
- Query comprehensions
Universal
- Projecting queries and images into a similarity space
- Query-by-image search and retrieval system
- Generating clusters of images for search results
- Automatic recommendation of vertical search engines
- Generating search results based on duplicate image detection
Temporal
- Determination of time dependency of search queries
- Forecasting time-dependent search queries
- Real time popularity prediction for events and queries
- Strategies for identifying anomalies in time-series data
- History answering for re-finding search results
- Using popularity data for ranking
Named Entities
- Named Entity transliteration using corporate corpra
- Entity-specific tuned searching
- Entity-specific search model
- Identifying synonyms of entities using web search
- Identifying synonyms of entities using a document collection
Geo Local
- System and method for providing a geographic search function
- Relevance improvements for implicit local queries
- System and method for providing a geographic search function
- Mining geographic knowledge using a location aware topic model
Relevance signals
Search by document type and relevance
Web Spam
Recommendation engine
Keyword search volume seasonality forecasting engine
TV Ads
Television system targeted advertising
Systemic
- Information Retrieval System
- System and method for facilitating full text searching utilizing inverted keyword indices
- System and method for generating normalized relevance measure for analysis of search results
- Indexing and searching of information including handler chaining
- System and method for customization of search results
- System for finding queries aiming at tail URLs
- Method and apparatus for web crawling
- Efficient retrieval algorithm by query term discrimination
Other
- Clustering question search results based on topic and focus
- Search results injected into client applications
- Method and system for summarizing a document
- Automatically Showing More Search Results
- Systems and methods for enhancing search query results
- Tag Ranking
- System and method for performing a search and a browse on a query
- Enhancing Search Query Results
- Knowledge extraction from online discussion forums
- Providing tools for navigational search query results
- Model Based Searching
- Generating search result summaries
- Presenting search results according to query domains
If you missed the round up yesterday, here are the Yahoo Search Patents 2010. On Monday we will have a look at the big dog in the park; Google. Until then, play safe!



