Machine learning and data mining

Researcher:
Prof. Ran El-Yaniv | Computer Science

Categories:

Information and Computer Science

The Technology

Semantic relatedness (SR) is about quantification of the intensity of how much two objects are related to each other. Most automatic SR valuation techniques rely on some kind of world or expert knowledge, which we term background knowledge.

The technology is based on a method of evaluating a semantic relatedness of terms. The method comprises providing a plurality of text segments and calculating, using a processor, a plurality of weights and a prevalence of a co-appearance of each of a plurality of pairs, and evaluating a semantic relatedness between members of each the pair.

Advantages

  • The proposed method exhibits good performance over a large and medium scale problems and excellent performance on small scale problems. In particular, it significantly outperforms the best supervised SR method, and distinctly superior to scores obtained by a plethora of unsupervised SR methods.

Applications and Opportunities

  • Targeted advertising
  • Information retrieval and web search
  • Automatic tagging
  • Translation
  • Summarizing textual documents
  • Text categorization
arrow Business Development Contacts
Motti Koren
Director of Business Development, Life Sciences