Some interesting news broke yesterday (on Search Engine Land) that I thought was worth noting here. While, on the surface, it probably means very little to the average search geek, to me this is huge.
According to the article, it seems that Anna Patterson has re-emerged at Google. Still not excited? Ok, a little background;
- Anna was once a Googler that left to start the (failed) search engine Cuil.
- She was previously responsible for large scale engine infrastructure
- She was the main architect of the phrase based IR series of patents.
That last one? This is what has me all giddy. Over the years Google has published a wide variety of patents relating to semantic analysis, but none had the scope and breadth of the series (of 9 patents) on phrase based indexing and retrieval. Why does that get me going? Because it was one of the most full featured semantic analysis approaches I’ve seen.
What’s this mean?
One can’t help but consider that this means Google must have truly seen value in Anna. Not only does she specialize in large scale engines, but brings these semantic golden nuggets to the table.
We also, not to beat up on a wounded equine, but it once more highlights why we can’t take approaches such as LDA to heart as THE method of choice as far as semantic analysis at Google is concerned (for more see; Google Rankings and LDA ). If you’re on the LDA bandwagon, you may want to re-consider that.
More on PaIR;
- What you need to know about phrase based optimization
- Phrase Based Optimization Resources
- Google granted a very Cuil patent
- Phrase-based personalization of searches in an information retrieval system
- Phrase Identification in an Information Retrieval System,
- Phrase-Based Generation of Document Descriptions,
- Phrase-Based Searching in an Information Retrieval System,
- Automatic Taxonomy Generation in Search Results Using Phrases,
- Phrase-based indexing in an information retrieval system
- Multiple index based information retrieval system
- Detecting spam documents in a phrase based information retrieval