Deep supervised hashing takes prominent advantages of low storage cost, high computational efficiency and good retrieval performance, which draws attention in the field large-scale image retrieval. However, similarity-preserving, quantization errors imbalanced data are still great challenges deep hashing. This paper proposes a pairwise similarity-preserving scheme to handle aforementioned probl...

We describe our participation in the WebCLEF 2007 task, targeted at snippet retrieval from web data. Our system ranks snippets based on a simple similarity-based centrality, inspired by the web page ranking algorithms. We experimented with retrieval units (sentences and paragraphs) and with the similarity functions used for centrality computations (word overlap and cosine similarity). We found ...

Similar web pages are pages that are about the same topic and of the same type. Sites about soccer clubs are related with all the soccer websites, but only similar with other soccer clubs. The goal of this research is to find an approach which, based on the textual content, can find similar pages given a page. The method used for this approach is a twofold method. The first task is trying to fi...

In this paper we consider the problem of retrieving the concepts of an ontology that are most relevant to a given textual query. In our setting the concepts are associated with textual fragments, such as labels, descriptions, and links to other relevant concepts. The main task to be solved is the definition of a similarity measure between the single text of the query and the set of texts associ...

Detecting levels of interest from speakers is a new problem in Spoken Dialog Understanding with significant impact on real world business applications. Previous work has focused on the analysis of traditional acoustic signals and shallow lexical features. In this paper, we present a novel hierarchical fusion learning model that takes feedback from previous multistream predictions of prominent s...

In this paper, we present a new probabilistic method for automatically extracting topic-specific strings in a text categorization context. The advantage of this method is twofold. First, it allows us to automatically point out the expressions characterizing a specific topic category for a potential knowledge modelling. Second, it contributes to improve categorization results by providing to the...

In the paper we present a method that allows an extraction of singleword terms for a specific domain. At the next stage these terms can be used as candidates for multi-word term extraction. The proposed method is based on comparison with general reference corpus using log-likelihood similarity. We also perform clustering of the extracted terms using k-means algorithm and cosine similarity measu...