Putting Search in Context: Using Dynamically-Weighted Information Fusion to Improve Search Results

The modern world is a highly connected one. Many diverse populations are utilizing the web as a learning tool. Often, the quest for information on the Internet begins by entering a search string into a search engine such as Google or Bing and then sifting through the multitude of results searching for information relevant to the user in his or her current situation. The web along with search engine technology has enabled people to have access to a wealth of information previously unavailable or out of reach of the average person. While the search process seems intuitive - enter a search string and "presto!" results are retrieved - constructing appropriate search strings and filtering results can be a complicated process, difficult for even the sophisticated user. Thus, the real issue that remains is perfecting the search tools and their underlying algorithms so that this search process will be simplified and personalized for the average user. As search engines are a typical user's interface to the World Wide Web, aiding the average user who possesses minimal query development skills in his or her quest for relevant, personalized results is an important goal. In addition, it is easy to see that mobile computing is also becoming pervasive in today's society. Use of personal digital assistants is common, and even cellular phones are Internet enabled. This trend is likely to continue, and this means that the user's context is likely to vary significantly as he or she uses these tools. Why not take advantage of context information to make the user's mobile computing experience simpler? USearch, an agent-based context-aware search system that fuses context information to improve search results is introduced here.