Methods based on machine learning have been proposed with certain advantages for TC (text categorization). However, it is still difficult to further increase the precision and understandability of categorization due to certain aspects of text itself. In this paper, we propose an architecture for TC by addressing domain ontology. Not only more effect and understandability of categorization are a...

This paper investigates an effective method of building domain-independent text generation system. In our English text generation system, Semantic Network is used as the internal Knowledge Representation. Nodes and links of the semantic network are classified according to word class and grammatical relations respectively. Generation results prove that the system works well and can generate cohe...

This paper examines the use of parallel and comparable corpora for automatic acquisition of semantics-extraction patterns. It presents a new method of the pattern extraction which takes advantage of parallel texts to “port” text mining solutions from a source language to a target language. It is shown that the technique can help in situations when the extraction procedure is to be applied in a ...

Most supervised language processing systems show a significant drop-off in performance when they are tested on text that comes from a domain significantly different from the domain of the training data. Semantic role labeling techniques are typically trained on newswire text, and in tests their performance on fiction is as much as 19% worse than their performance on newswire text. We investigat...

In this paper we will present a system that is able to perform cooperative information retrieval actions over a text knowledge base. The knowledge base is composed by four levels: Interaction, Domain, Information Retrieval and Text. The interaction level is responsible for the dialogue management, including the inference of attitudes. The domain level is composed by rules encoding knowledge abo...

In this paper we will present a system that is able to perform cooperative information retrieval actions over a text knowledge base. The knowledge base is composed by four levels: Interaction, Domain, Information Retrieval and Text. The interaction level is responsible for the dialogue management, including the inference of attitudes. The domain level is composed by rules encoding knowledge abo...

We describe the creation of a corpus that supports a real-world hierarchical text categorization task in the domain of electronic rulemaking (eRulemaking). Features of the task and of the eRulemaking domain engender both a non-traditional text categorization corpus and a correspondingly difficult machine learning task. Interannotator agreement results are presented for a group of six annotators...