Victor Shafran

Learn More
Schema and ontology matching is a process of establishing correspondences between schema attributes and ontology concepts, for the purpose of data integration. Various commercial and academic tools have been developed to support this task. These tools provide impressive results on some datasets. However , as the matching is inherently uncertain, the(More)
Schema matching in uncertain environments faces several challenges, among them the identification of complex correspondences. In this paper, we present a method to address this challenge based on top-k matchings, i.e., a set of matchings comprising only 1: 1 correspondences derived by common matchers. We propose the unified top-k match graph and define a(More)
Given a schema and a set of concepts, representative of entities in the domain of discourse, schema cover defines correspondences between concepts and parts of the schema. Schema cover aims at interpreting the schema in terms of concepts and thus, vastly simplifying the task of schema integration. In this work we investigate two properties of schema cover,(More)
Coverage is a task, where a robot is to move about a given a target area until every point in it is visited. Many efficient coverage algorithms cannot be used in practice, because they assume accurate movements by the robot; unfortunately, real robots have navigational errors. A standard costly solution is to utilize a robot that continuously localizes, so(More)
Sentiment analysis (SA) is a central component in industrial text analytic solutions enabling the automatic detection of customers opinions regarding products and services. However, even though state-of-the-art techniques achieve a very high level of accuracy on written texts, these techniques perform significantly worse on spoken conversation which is(More)
This thesis has two parts. The first part presents a hybrid approach to the cov-erge problem under dead reckoning errors. Coverage is a canonical robotics task, where single or multiple robots are given a target work area, and move about the area until every point in the area is visited by the robots. There are many efficient exact-motion coverage(More)
Matchmaking is the process of introducing two or more agents to each other. Current matchmaking techniques are unidirectional and fail to address large-scale and highly dynamic systems with time constraints. We propose a new distributed technique which scales well, and still maintains relatively low matchmaking time and communication overhead. Our technique(More)
  • 1