Topic Discovery
In addition to basic search results, IT.com provides in response to each query a list of topics that are related to what the searcher is looking for. We use imputed topic analysis to find topics that naturally occur within the data. Then, when a searcher enters a query, we're able to discover what topics are relevant to that particular search, and list them alongside the search results. This allows them to organize their results, suggests alternate lines of inquiry, and provides a view to the type of topics addressed by the corpus as a whole. We call this process "topic discovery".
Unlike "clustering" technology, which attempts to organize the results of a particular search, IT.com's imputed topic analysis is able to pre-analyze a collection of documents and discover topics, prior to user queries. In this way, we're able to take advantage of a taxonomy that occurs naturally within the corpus, not just the first couple of hundred results.
Using this technique, we are able at search time to suggest existing topics that are strongly related to the user's search. This allows the user to:
- Find search results that might use slightly different, but equivalent terminology to their search
- Organize the results of their query into topics that reflect the contents of the document collection
- Search within a topic, refining their search results to that topic's focus
- Subscribe to a topic, and be alerted to new topical content that enters the system
- Browse the corpus via an organized, automatically-derived "taxonomy".