Today, companies have access to unprecedented volumes of data that they can convert into strategic business tools, from supplying the insights they need to be competitive in their market, to the right information at the right time to support decision making. To be able to benefit from what this data has to offer requires technologies able to fill in the information gaps: Enter text analytics.
What is text analytics? According to several reports, it represents a growing market that is expected to grow to over $10 billion by 2020. In terms of what text analytics does, there are a lot of misconceptions. To many business users, text analytics is a box where unstructured text goes in and structured information magically comes out (sounds like search, right?). Not exactly. Through the use of a complex taxonomy and algorithms, text analytics is essentially about making human information (text, documents, language) understandable by computers and a way to discover value in text. Text Analytics is a process that can analyze any type of unstructured text, extracting high-quality and relevant information that can drive further analysis and strategic decision making.
Traditional analysis technologies are not able to effectively handle unstructured data because they merely manage lists organized in databases, and, equally important, without understanding the meaning of the terms, they cannot find the precise information contained within the text. (like search)
Text analytics at work
At work, this process allows companies to read between the lines, understanding meaning and context, and recognizing patterns or complex relationships that may not be immediately visible to the human eye, from online, social and enterprise text.
Let’s look at a practical application of text analytics in a high volume industry: pharmaceutical. Here, it is common for the FDA to request additional information related to specific tests conducted during the experimentation phase. To comply, scientists and knowledge workers must sort through huge volumes of documents to retrieve the proper documentation.
Expert System implemented a discovery application that utilizes automatic extraction and tagging of all relevant concepts around human pathology. This ensures the highest level of precision and recall in search, and allowed the teams to respond quickly to FDA requests. Applying the same techniques to the FDA website, the application enables automatic identification of all FDA-issued approvals for company products. This allows automatic routing and activation of a workflow to ensure faster turnaround of information (and faster go to market). Text analytics does this through the analysis of the content, using a process known as entity extraction to extract references to people, products, locations, and related terms and concepts. It then groups similar pieces of information together, and starts mapping relationships between the extracted references.
So, text analytics may not be a magic box, but a business partner? Definitely.
