Back to results
US 2020/0026799 A1Published application
Dictionary Editing System Integrated With Text Mining
AssigneeInternational Business Machines Corporation
InventorsSusumu Fukuda, Kenta Watanabe, Shunsuke Ishikawa, Takashi Fukuda
Life of this patent
Estimating the term…
Abstract
Embodiments are directed to a system, computer program product, and method for dynamic facet dictionary management. As one or more annotations are applied to a document collection, electronic text and associated facets are identified. Additional facets and facet values are identified and selectively applied to a knowledge base. A dictionary comprised of facets and associated facet values is constructed from the selective application. Application of the dictionary to the knowledge base identifies and returns a targeted document collection. Accordingly, facet mining and dictionary construction are dynamically applied to the knowledge base.
Claims24
A system comprising:
- a processing unit operatively coupled to memory;
- a functional unit in communication with the processing unit, the functional unit having tools to support text mining and dictionary construction, the tools to integrate an annotation with text mining, including application of the annotation to a knowledge base, the tools comprising:
- a document manager to integrate an annotation with the text mining, including the document manager to apply the annotation to the knowledge base and return a first collection of documents from the knowledge base based on the annotation;
- a text manager operatively coupled to the document manager, the text manager to:
- extract two or more words from the returned first collection of documents, the extracted two or more words associated with applied the annotation;
- select at least one word from the extracted two or more words, and designate the selected word as a facet;
- and
- selectively add a facet value to the facet, the facet value selected from one or more linguistic terms identified from the first collection of documents returned by the document manager, the first collection determined to be related to the facet;
- a facet manager operatively coupled to the text manager, the facet manager to dynamically execute a statistical analysis of the selectively added facet value, the analysis including calculation of an approximate value for the temporary addition of the facet value;
- a dictionary manager operatively coupled to the facet manager, the dictionary manager to:
- construct a first temporary dictionary comprised of the annotation, the facet, and one or more of the temporary added facet values;
- and
- selectively apply the constructed first temporary dictionary to the first collection response to the statistical analysis;
- and
- a targeted list of documents returned from the selective application of the first temporary dictionary to the first collection.
12 more elements
A computer program product for text mining, the computer program product comprising a computer readable storage device having program code embodied therewith, the program code executable by a processor to:
- integrate an annotation with a text mining system, including the program code to apply the annotation to a knowledge base and return a first collection of documents from the knowledge base based on the annotation;
- extract two or more words from the returned first collection of documents, the extracted two or more words associated with applied the annotation;
- select at least one word from the extracted two or more words, and designate the selected word as a facet;
- temporarily add a facet value to the facet, the facet value selected from one or more linguistic terms identified from the returned first collection of documents, the first collection determined to be related to the facet;
- dynamically execute a statistical analysis of the selectively added facet value, the analysis including calculation of an approximate value for the temporary addition of the facet value;
- construct a first temporary dictionary comprised of the annotation, the facet, and one or more of the temporary added facet values;
- and
- selectively apply the constructed first temporary dictionary to the first collection response to the statistical analysis;
- and
- a targeted list of documents returned from the selective application of the first temporary dictionary to the first collection.
7 more elements
A method comprising:
- integrating an annotation with a text mining system, including applying the annotation to a knowledge base, and returning a first collection of documents from the knowledge base based on the annotation;
- extracting two or more words from the returned first collection of documents, the extracted two or more words associated with the applied annotation;
- selecting at least one word from the extracted two or more words, and designating the selected word as a facet;
- temporarily adding a facet value to the facet, the facet value selected from one or more linguistic terms identified from the returned first collection of documents determined to be related to the facet;
- dynamically executing a statistical analysis of the selectively added facet value, the analysis calculating an approximate value for the temporary facet value addition;
- constructing a first temporary dictionary comprised of the annotation, the facet, and one or more of the temporary added facet values;
- selectively applying the constructed first temporary dictionary to the first collection response to the statistical analysis;
- and
- a targeted list of documents returned from the selective application of the first temporary dictionary to the first collection.
6 more elements
A system comprising:
- a processing unit operatively coupled to memory;
- a functional unit in communication with the processing unit, the functional unit having tools to support text mining and dynamic dictionary construction, the tools comprising:
- a document manager to apply an annotation to a knowledge base, the application to return a first collection of documents based on the applied annotation;
- a text manager operatively coupled to the document manager, the text manager to:
- identify one or more facets in the first collection, the identified one or more facets linguistically related to the applied annotation;
- for each identified one or more facets, the text manager to:
- extract two or more words from the returned first collection of documents, each extracted word linguistically associated with the identified facet in the first collection;
- responsive to selection of at least one word from the extracted two or more words, the text manager to:
- designate the selected at least one word as a facet value;
- temporarily assign the facet value to the facet;
- in real-time, statistically analyze a strength of a linguistic relationship of the designated facet value to the identified facet;
- and
- selectively add the facet value to the facet responsive to the statistical analysis;
- and
- a dictionary created from storing the identified facet and selectively added facet value.
12 more elements
A method comprising:
- applying an annotation to a knowledge base and returning a first collection of documents based on the applied annotation;
- identifying one or more facets in the first collection, the identified one or more facets linguistically related to the applied annotation;
- for each identified one or more facets:
- extracting two or more words from the returned first collection of documents, each extracted word linguistically associated with the identified facet in the first collection;
- selecting at least one word from the extracted two or more words;
- designating the selected at least one word as a facet value;
- and
- temporarily assigning the facet value to the facet;
- in real-time, statistically analyzing the a strength of a linguistic relationship of the designated facet value to the identified facet
- selectively adding the facet value to the facet responsive to the statistical analysis;
- and
- creating a dictionary storing the identified facet and selectively added facet value.
9 more elements
Dependencies are inferred from the claim text.