Marco Cammisa

Semantic Kernels for Text Classification Based on Topological Measures of Feature Similarity

In this paper we propose a new approach to the design of semantic smoothing kernels for text classification. These kernels implicitly encode a superconcept expansion in a semantic network using well-known measures of term similarity. The experimental evaluation on two different datasets indicates that our approach consistently improves performance in situations of little training data and data sparseness.

Effective use of WordNet Semantics via Kernel-Based Learning

Research on document similarity has shown that complex representations are not more accurate than the simple bag-of-words. Term clustering, e.g. using latent semantic indexing, word co-occurrences or synonym relations using a word ontology have been shown not very effective. In particular, when to extend the similarity function external prior knowledge is used, e.g. WordNet, the retrieval system decreases its performance. The critical issues here are methods and conditions to integrate such knowledge. In this paper we propose kernel functions to add prior knowledge to learning algorithms for document classification. Such kernels use a term similarity measure based on the WordNet hierarchy. The kernel trick is used to implement such space in a balanced and statistically coherent way. Cross-validation results show the benefit of the approach for the Support Vector Machines when few training data is available.

RitroveRAI: a web application for semantic indexing and hyperlinking of multimedia news

In this paper, a system, RitroveRAI, addressing the general problem of enriching a multimedia news stream with semantic metadata is presented. News metadata here are explicitly derived from transcribed sentences or implicitly expressed into a topical category automatically detected. The enrichment process is accomplished by searching the same news expressed by different agencies reachable over the Web. Metadata extraction from the alternative sources (i.e. Web pages) is similarly applied and finally integration of the sources (according to some heuristic of pertinence) is carried out. Performance evaluation of the current system prototype has been carried out on a large scale. It confirms the viability of the RitroveRAI approach for realistic (i.e. 24 hours) applications and continuous monitoring and metadata extraction from multimedia news data.

A semantic kernel to exploit linguistic knowledge

Improving accuracy in Information Retrieval tasks via semantic information is a complex problem characterized by three main aspects: the document representation model, the similarity estimation metric and the inductive algorithm. In this paper an original kernel function sensitive to external semantic knowledge is defined as a document similarity model. This semantic kernel was tested over a text categorization task, under critical learning conditions (i.e. poor training data). The results of cross-validation experiments suggest that the proposed kernel function can be used as a general model of document similarity for IR tasks.