VortexT’s Recommender Content Search Engine immediately focuses analysts on the most relevant data—no more wasting time wading through extraneous information and refining keyword searches.
Unstructured textual data, growing at exponential rates, account for over 80% of data available to organizations…data that is not being leveraged due to human and technological constraints. As data populations grow, keyword searches, no matter how sophisticated, become more and more ineffective. VortexT’s Recommender uses the entire content of example documents or unlimited lists of terms/phrases to create a profile as the basis for searches. The full content (i.e., every word) in the profile is compared to the full content (i.e., every word) in each of the ingested target data files to determine their similarity, thereby incorporating the context and concepts within the data into the search and analysis.
Optional stop word lists further concentrate content search results on the most relevant information, which is delivered to the analyst individually rank-ordered against each document used to create the profile. Analysts are provided a full toolset, such as highlights, word incidence, and frequency charts, to further support their evaluation and decision-making process.
Once again, the computational efficiencies of VortexT’s linear algorithms, combined with more efficient analysis, allow users to more accurately and efficiently identify targeted information and discover new information through evaluation of larger target data sets.
Profile Documents (PDs) are the documents the user selects that provide the best and most complete description of the information he/she is interested in retrieving and reading. PDs can range from a page to hundreds of pages. We feed the PDs into VortexT, which analyzes it and determines the group of the most important terms in the document. Then, hundreds of thousands or even millions of documents can be ingested into VortexT to undergo a similar process and are then matched to the PDs.