Text mining is the process of transforming unstructured text into structured data for easy analysis. It uses natural
language processing tools to interpret the human language and process text documents automatically. The extracted
insights provide a valuable evaluation tool, which can provide systematic feedbacks and can drive machine-learning
algorithms. We have applied these techniques to the SPIE proceeding papers, dedicated to the x-ray optics with special
regards to the Optics for EUV, X-Ray, and Gamma-Ray Astronomy conference text data corpus, accessible through
the ADS website. In particular, we worked on the collection of text formed by titles, abstracts, authors and affiliations
to extract patterns, correlations and knowledge.
|