Search relevance using messages of a messaging platform

#Traditional search techniques depend on a variety of signals to identify and rank search results. These signals can involve low latency query-dependent analysis as well as higher latency query-independent analysis. When dealing with a large corpus of documents, it is important to select signals that (1) can be efficiently analyzed and (2) have a high relevance correlation with the desired results. Thus, a fast and efficient signal is useless if it produces non-relevant results. Conversely, a high-relevance signal can be infeasible for low-latency applications.

#One example of a common relevance signal used by modern search engines is hyperlink anchor text. Hyperlink anchor text is the visible and selectable text of a hyperlink, usually displayed on a referring web page. Generally, hyperlink anchor text includes concise, descriptive, and contextually relevant terms, which can be useful in relevance scoring, when available. Though undoubtedly useful in a variety of applications, hyperlink anchor text as a relevance signal has some notable shortcomings. Specifically, hyperlink anchor text can be relatively slow to acquire since new web pages require time to accumulate inbound links. This type of delay can seriously hamper a time-critical or real-time analysis. Furthermore, hyperlink anchor text tends to consist primarily of descriptive language (i.e., a title), and can be a poor indicator of sentiment.

#Hyperlink anchor text is just one example of a common relevance signal, and is typically used for web search. Search engines use a variety of relevance signals, which are weighted differently according to the application-specific requirements involved. For example, a real-time search engine can have entirely different requirements from those of a traditional search engine and may put a greater emphasis on timeliness as a search criterion. Regardless of the specific application, the right relevance signals must be selected, and the right combination of those relevance signals must be applied to produce meaningful, relevant results.

#Embodiments of the present invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements.

#FIG. 1 shows a schematic diagram of a system, in accordance with one or more embodiments of the invention.

#FIG. 2 shows an example depiction of a conversation graph, in accordance with one or more embodiments of the invention.

#Specific embodiments of the invention will now be described in detail with reference to the accompanying figures. Like elements in the various figures are denoted by like reference numerals for consistency.

#In the following detailed description of embodiments of the invention, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to one of ordinary skill in the art that the invention may be practiced without these specific details. In other instances, well-known features have not been described in detail to avoid unnecessarily complicating the description.

#In general, embodiments of the invention provide a method and system for searching documents. Messages including a reference to a document (e.g., a web page) are identified and message text is identified within the messages. The message text is then used as a relevance signal for purposes of identifying relevant documents in response to a search request.