This course covers the introduction into the basic ideas of data mining and shows how data mining can be used to classify text documents. This can for example be useful for the automatic routing of e-mail messages to a group of people (e.g. for customer support) but can also be the core of a personalized news agent. Finally, this course lead to the analysis of web documents and concentrates on forum and blog sites: the statements of users about products and brands are freely available and build the base of the analysis of sentiments of your (potential) customers.

This course shows how you could know what your customers and potential customers think about your products in a very timely fashion from information provided by them and are freely available in the web. This course enables you to quickly build mash-ups to extract and integrate information from various sources on the web and to automatically crawl and categorize web pages using the latest text mining technologies. All theoretical ideas are transferred to the corresponding RapidMiner processes which can then be adapted to your own tasks of sentiment analysis, opinion mining, or market research.

Details

• Course ID: 1103
• Number of days: 2 days
• Location: Dortmund, Germany
• Target audience: decision makers, data mining users, analysts, developers, administrators
• Previous knowledge: basic knowledge of computer programs
• Methods: lectures, discussions, individual and group work, exercises on realistic data. Participants may introduce own work and project specific questions in order to find particular solutions together with the trainer and other participants.
• Content: this course includes a compact introduction into the foundations of data mining and into the software RapidMiner. It addresses beginners and intermediate learners. Topics of this course are
  • Basics of Data Mining with RapidMiner
  • Introduction into Data Mining on the web and on texts
  • Connecting a web crawler to RapidMiner
  • Preprocessing of textual data: tokenizing, tagging, term frequencies, document frequencies, TFIDF
  • Machine Learning methods for text mining: Naive Bayes and Support Vector Machines
  • Automatic text classification and clustering
  • Classifying web pages or blogs or discussion entries according to sentiments or opinions
  • Segmenting texts into different opinions
  • Prediction of sentiments and opinions from given text documents
  • Aggregation and statistics of sentiments
  • Evaluating the quality of aggregated sentiments
  Extensive exercises on different data sets will be performed.

Prices

Value added tax (VAT) may have to be added to these prices.