Coreference resolution is the process of identifying when two noun phrases (NP) refer to the same entity. This dissertation makes two main contributions to computational coreference resolution. 
First, this work contributes a new method for recognizing when an NP is anaphoric. Most pronouns have an antecedent, but many definite noun phrases do not. I present an unsupervised model for learning nonanaphoric definite NPs from a text collection, and I show that it learns lists of these noun phrases with good accuracy. Recall of these NPs increases from 43% to 79%. I also demonstrate that using these lists to filter nonanaphoric definite NPs prior to coreference resolution provides a mechanism for effecting a recall/precision tradeoff. In two distinct testing domains, recall is traded for precision, leading to precision increases from 60% to 73% and from 68% to 82%. 
Second, traditional approaches to coreference resolution typically select the most appropriate antecedent by recognizing word similarity, proximity, and agreement in number, gender, and semantic class. This work contributes a new source of evidence that focuses on the roles that an anaphor and antecedent play in particular events or relationships. I show that using contextual role knowledge as part of the coreference resolution process increases the number of anaphors that can be resolved, and I demonstrate an unsupervised method for acquiring contextual role knowledge that does not require an annotated training corpus. A probabilistic model based on the Dempster-Shafer model of evidence is used to incorporate contextual role knowledge with traditional evidence sources. Among the advantages of this model is the capability to assign evidence to a set of candidates when a knowledge source is unable to distinguish among them. In the two testing domains, the F-measure of anaphor/antecedent pairs increases from 0.57 to 0.61 and from 0.57 to 0.63. Recall increases from 46% to 53% and from 42% to 51% with only minor reductions in precision.

Acquisition and application of contextual role knowledge for coreference resolution

Technologies are described herein for coreference resolution in an ambiguity-sensitive natural language processing system. Techniques for integrating reference resolution functionality into a natural language processing system can processes documents to be indexed within an information search and retrieval system. Ambiguity awareness features, as well as ambiguity resolution functionality, can operate in coordination with coreference resolution. Annotation of coreference entities, as well as ambiguous interpretations, can be supported by in-line markup within text content or by external entity maps. Information expressed within documents can be formally organized in terms of facts, or relationships between entities in the text. Expansion can support applying multiple aliases, or ambiguities, to an entity being indexed so that all of the possibly references or interpretations for that entity are captured into the index. Alternative stored descriptions can support retrieval of a fact by either the original description or a coreferential description.

Coreference resolution in an ambiguity-sensitive natural language processing system

A system for analyzing language using supervised learning method. The system extracts portions matching the structures of problem expressions from a raw corpus that is not supplemented with analysis information, then converts the extracted portions corresponding to the problem expressions into supervised data including problems and solutions and stores in the data storage. The system extracts sets of solutions and features from the supervised data stored in the data storage, carries out machine learning processing using the sets and stores learned results as to what kind of solution is the most straightforward for which feature in the learning results database. The system then extracts sets of features from the inputting object data, extrapolates analysis information showing the most optimum for a certain feature, from the sets of features based on the learning results database.

System and method for analyzing language using supervised machine learning method

An embodiment of the present invention provides a method for automatically translating text. First, a conceptual representation space is generated based on source-language documents and target-language documents, wherein respective terms from the source-language and target-language documents have a representation in the conceptual representation space. Second, a new source-language document is represented in the conceptual representation space, wherein a subset of terms in the new source-language document is represented in the conceptual representation space, such that each term in the subset has a representation in the conceptual representation space. Then, a term in the new source-language document is automatically translated into a corresponding target-language term based on a similarity between the representation of the term and the representation of the corresponding target-language term.

Machine translation using vector space representations

The disclosed invention performs a set of operations on an email to analyze the text and generate a coherent summary. Email summaries are generated by applying a coherence layer after a ranking process. Analyzing how sentences relate to each other via discourse markers and other linguistic devices aids in enhancing coherence of the email summaries. Output summaries are more coherent and easier to understand because they mimic the flow of ideas contained in the original message instead of merely being a collection of extracted sentences. Tasks may also be extracted from the text of the email to assist users in keeping track of tasks that they receive via email.

Method and system for summarizing emails and extracting tasks

An anaphora analyzing apparatus is disclosed for automatically estimating an anaphora referential relation or an antecedent of a noun for use in a natural language sentence. A storage unit stores analyzed results outputted from an analyzer, and an antecedent candidate generator detects a target component required for anaphora analysis in accordance with the current analyzed results and the past analyzed results stored in the storage unit, and generates antecedent candidates corresponding to the target component. Then, a candidate rejecting section rejects unnecessary candidates having no potential for anaphora referential relation among the antecedent candidates by using a predetermined rejecting criterion, and outputs the remaining antecedent candidates. Further, a preference giving section calculates a predetermined estimated value for each of the remaining antecedent candidates by referring to an information table including predetermined estimation information obtained from a predetermined training tagged corpus, and gives the antecedent candidates preference in accordance with the calculated estimated value. Finally, a candidate deciding section decides a predetermined number of antecedent candidates based on a given preference in accordance with the preferenced antecedent candidates.

Anaphora analyzing apparatus provided with antecedent candidate rejecting means using candidate rejecting decision tree

The device has an analyzer (1) for analyzing an entered natural speech file and outputting the analysis results, a memory medium (11) for storing the output analysis results, a predecessor candidate generator (2) for detecting a target component in the entered file required for the anaphora analysis in accordance with the current analysis results and the preceding stored analysis results and for generating the predecessor candidates corresponding. to the target components, a candidate rejection device (3), a preference allocation device (4) and a candidate decision device (5) for determining and outputting a defined number of candidates with preferences.

Anaphora analysis device has analyzer, analysis result memory, predecessor candidate generator, candidate rejection device, preference allocation device, candidate decision device

PROBLEM TO BE SOLVED: To provide a correspondence analyzing device capable of analyzing correspondence with higher accuracy than conventional techniques. SOLUTION: A correspondence candidate generation part 2 detects an object element requiring correspondence analysis on the basis of the analysis result at present and the analysis result in the past stored in an analysis result memory 11 and generates a correspondence candidate corresponding to that object element, and a candidate dismissing part 3 dismisses unwanted candidates having no possibility in a corresponding relation from the generated correspondence candidates by using a dismissing reference as a determination tree provided by using the method of mechanical learning on the basis of a corpus with a tag for learning to which word information is given to each word, and outputs the remaining correspondence candidates. Next, a priority imparting part 4 calculates a prescribed evaluation value for each remaining correspondence candidate while referring to an information table including prescribed evaluation information obtained from the prescribed corpus with the tag for learning and outputs the correspondence candidates after imparting priority on the basis of the calculated evaluation values, and a candidate determining part 5 determines and outputs a prescribed number of correspondence candidates on the basis of the imparted priority.

Michael Paul

Papers

Anaphora analyzing apparatus provided with antecedent candidate rejecting means using candidate rejecting decision tree

Anaphora analysis device has analyzer, analysis result memory, predecessor candidate generator, candidate rejection device, preference allocation device, candidate decision device

Correspondence analyzing device