Bayesian networks have been known as a powerful knowledge representation model and a computational architecture for reasoning under uncertainty. However, there are still some obstacles in applying them for practical applications. Among them are learning parameters, or the conditional probability tables from data. Much work in the field focuses on handling incomplete data, i.e. data with missing values. In this paper, we propose an approach for learning parameters of Bayesian networks from soft data, the values of which may be imprecise or uncertain, given by beliefs or confidence scales. Theoretical justification of the approach is given, and experimental results demonstrated the effectiveness of our approach.

To explore the fusion of unstructured text data, the concept of entity networks has been proposed in many systems to facilitate analysis of entity relationship. Using appropriate visualization, entity networks could illustrate the relations of extracted entities. However, such a system generates too many entities and links from a single document, and many of them are trivial. With a complicated entity network, it is difficult to grasp the key points of the text. Also it is less effective to combine complicated entity networks across time. Using novel and highly efficient algorithms for shortest-path estimating and tracing, this paper proposes a simplification scheme for entity networks, which has two main functions. One is to divide an entity network into segments, with each segment representing a meaningful event. Another is to identify the topic by extracting primary nodes and the linking paths. With the simplified network, we could observe the overall structure, and get the key concepts and their relationship. The examples in this paper demonstrate the purport and effectiveness of the simplification scheme.

In the domain of fusion for passive signals, the confidence degrees for conversations and communication networks are critical to obtain the precise belief degree for higher level information. In this paper, we first examine the characteristics of messages, emitters, conversation, and communication network. By defining several temporal functions, we then propose the formulation of association degree, termed as homology degree, for clustering messages into emitters, and present mathematical formulation to compute the confidence degree for any possible conversation between a pair of emitters. Subsequently, by using fuzzy matrix operations, we introduce the incidence matrix for a set of ongoing conversations, and design an iterative fuzzy matrix operation to deduce the confidence matrix for a communication network. Also, by using Dempster-Shafer combination rule, we describe a method to combine two confidence matrices for a network from different sources, or at different time steps. Finally, we provide an example to show the purport and effectiveness of our methods.

This paper presents a set of novel and efficient algorithms for fusion of emitters from multiple sensors, where a signal from one sensor system could either be a bearing line or a fix. This comprehensive solution includes fusion algorithms for multiple fixes, multiple bearing lines, and fix with bearing. Testing using a high-fidelity simulation system demonstrated that the algorithms are able to reduce the location error significantly

The understanding of Semantic Web documents is built upon ontologies that define concepts and relationships of data. Hence, the correctness of ontologies is vital. Ontology reasoners such as RACER and FaCT have been developed to reason ontologies with a high degree of automation. However, complex ontology-related properties may not be expressible within the current web ontology languages, consequently they may not be checkable by RACER and FaCT. We propose to use the software engineering techniques and tools, i.e., Z/EVES and Alloy Analyzer, to complement the ontology tools for checking Semantic Web documents.In this approach, Z/EVES is first applied to remove trivial syntax and type errors of the ontologies. Next, RACER is used to identify any ontological inconsistencies, whose origins can be traced by Alloy Analyzer. Finally Z/EVES is used again to express complex ontology-related properties and reveal errors beyond the modeling capabilities of the current web ontology languages. We have successfully applied this approach to checking a set of military plan ontologies.

A probabilistic method is used for word sense disambiguation where the features taken are the surrounding six words. As their surface forms are used, no syntactic or semantic analysis is required. Despite its simplicity, this method is able to disambiguate the noun interest accurately. Using the common data set of (Bruce & Wiebe 94), we have obtained an average accuracy of 86.6% compared with their reported figure of 78%. This portable technique can be applied to the task of English article selection. This problem arises from machine translation of any source language without article to English. Using texts from the Wall Street Journal, we achieved an overall accuracy of 83.1% for the 1,500 most commonly used head nouns.

In this paper, we present a new approach for word sense disambiguation (WSD) using an exemplar-based learning algorithm.This approach integrates a diverse set of knowledge sources to disambiguate word sense, including part of speech of neighboring words, morphological form, the unordered set of surrounding words, local collocations, and verb-object syntactic relation.We tested our WSD program, named LEXAS, on both a common data set used in previous work, as well as on a large sense-tagged corpus that we separately constructed.LEXAS achieves a higher accuracy on the common data set, and performs better than the most frequent heuristic on the highly ambiguous words in the large corpus tagged with the refined senses of WoRDNET.