This week in The Trend Point, we have spotted quite a few articles related to text analysis. As semantic technologies improve, this process is becoming more and more essential to ROI and increased productivity in the workplace.
Semantic search enables better relevancy and results. According to “Evolution of the Web and Semantic Search“:
Everything in it has meaning and every connection adds further potential layers of inference, which then lead to more subtle nuanced meaning…It’s machine driven and computer code generated yet it feels like it’s awakening because, for once, the connections make sense… When data is being used to make decisions, understand trends and power developments and you and I are part of that data then it allows us to affect it. In small ways, perhaps, but important ones nonetheless.
Then, from “Content Analysis Methodology Paper A Valuable Source of Information” we learned that even academia is tapping into the power of semantic technology:
We provide a guide to this exciting new area of research and show how, in many instances, the methods have already obtained part of their promise. But there are pitfalls to using automated methods—they are no substitute for careful thought and close reading and require extensive and problem-specific validation. We survey a wide range of new methods, provide guidance on how to validate the output of the models, and clarify misconceptions and errors in the literature. To conclude, we argue that for automated text methods to become a standard tool for political scientists, methodologists must contribute new methods and new methods of validation.
As far as making sense out of data, visualizing normalized textual data as a graph teasing out key concepts and contextual clusters is one way discussed in a recent article on The Trend Point. “Text Network Analysis Method from Nodus Labs” tells us:
Any text can be represented as a network. At a basic level, the words, or the concepts are the nodes, and their relations are the edges of the network. Once a text is represented as a network, a wide range of tools from network and graph analysis can be used to perform quantitative analysis and categorization of textual data, detect communities of closely related concepts, identify the most influential concepts that produce meaning, and perform comparative analysis of several texts.
Making sense out of raw data and churning it into insights has been the name of the game for Unified Information Access at Sinequa for a while. This technology incorporates the power of a library of connectors, over one hundred strong. Text analysis from multiple file types has never been easier. Yet, the technology is tuned to customer environments in order to take in specific dictionaries, taxonomies and companies’ ideas of what is “close” in semantic terms. Complexity of semantic analysis does not disappear by “magic”, but it is dealt with behind the scenes such that semantic search become easy for end users. And semantic search vendors work hard – at least they should! – in order to make life of the semantic tuning experts as easy as possible as well, while maintaining the freedom and flexibility they need.
Jane Smith, April 17, 2012