Abstract “SeMoDesk” is an approach to implement the Semantic Desktop on mobile devices. The idea is to allow users to manage their personal information space using personal ontologies. In this paper, we are presenting our solution to improve location-awareness in this scenario. We have designed a location and sensor ontology as an extension to the personal ontology. This ontology is then used to retrieve relevant resources according to the current user context. For this purpose we have designed a resource recommendation function that is utilizing the ontology graph to find other related resources such as persons or documents. We describe the design, implementation and test of an extended SeMoDesk with a focus on the integration of a RFID infrastructure for indoor location-awareness. Furthermore, we have implemented a method to display current resources and points-of-interests in the user vicinity on a map. To learn relevant user locations, we have designed and implemented a solution based on a time-based clustering algorithm. This method has been evaluated with good results in our scenario. Read More Read Less

Abstract The Web is a huge repository of information in the form of text, image, audio, and video. People use search engines, such as Google, Yahoo!, Bing, etc, to discover resources from this huge repository. Originally these general purpose search engines are designed and optimized for English language. They fell short when they are used for locating web resources of other languages such as Amharic. This is mainly due to the specific features of the language that are not considered by those search engines. Amharic, which is a family of Semitic languages, is the working language of the federal government of Ethiopia. Currently, there are significant numbers of Amharic documents on the Web. In this work, we analyzed the specific features of Amharic language, designed a general architecture for Amharic Search Engine, developed the necessary algorithm to realize it and implemented the same for searching Amharic language web documents. The result of the work has become to be a complete language specific search engine that has a crawler, an indexer and a query engine component that are optimized for the language they are designed, Amharic language. Read More Read Less

Abstract In this paper, we present a validation approach of detected alignment links between dialog transcript and discussed documents, in the context of a multimodal document alignment framework of multimedia events (meetings and lectures). The validation approach consists in an entailment process of the detected alignment links. This entailment process exploits several features, from the structural level of aligned documents to the linguistic level of their tokens. The implemented entailment strategies were evaluated on several multimodal corpora. The obtained results prove that the choice of the relevant entailment strategy depends on the types of documents that are available in the corpus, on their content, and also on the nature of the corpus. Read More Read Less

Abstract The aim of the paper is to present and apply a QUery EXpansion MEthod called QUEXME while querying the Euro-Mediterranean Information System (EMWIS) on know-how in the Water sector. EMWIS provides a strategic tool for exchanging information and knowledge in the water sector between and within the Euro Mediterranean partnership countries (www.emwis.net). Information retrieval on the web or through some cooperation of information sources or some general knowledge bases is a complex process and a great challenge with the emergence of the semantic web. The aim of the query expansion method is to help and guide users to build their requests giving them some usually related terms close to their queries. Information retrieval in EMWIS is based on the use of a thesaurus to query the information system and to find relevant documents on some specific topics in the water sector. This thesaurus can be viewed as a light-weight web ontology. It is multilingual. This paper proposes an experimentation of our query expansion method within the framework of the EMWIS information system. Read More Read Less

Abstract Local features are widely adopted to describe visual information in tasks for image registration and matching. Nowadays the most used and studied local feature is SIFT (Scale Invariant Feature Transform)[1] since it assures a powerful local description and the invariance when little changes in the viewpoint occur. We propose a feature that is based on SIFT features and tends to capture larger image areas in images and can be used for semantic based task. These features are called bi-SIFT for their resemblance with textual bigrams. We tested the capability of the proposed representation with Corel data-set and publicly available image dataset. In particular we calculated the most representatives features through a clusterization process and used these value according to the visual terms paradigm. Experiments on the representation of sets of images with the proposed representation are shown. Results appear to be encouraging. Read More Read Less