Abstract: It is an object to highly accurately perform semantic relationship extraction from text data by performing supervised learning of multiple classes using an existing thesaurus as a correct answer. Concerning any pair of words in a text, a plurality of kinds of similarities are calculated and a feature vector including the similarities as elements is generated. A label indicating a classification of a semantic relationship is given to pairs of words on the basis of the thesaurus. Data for semantic relationship identification is learned as an identification problem of multiple classes from the feature vector and the label. Identification of an inter-semantic relationship of two words is performed according to the data for semantic relationship identification.

Abstract: There is disclosed an interface enabling easy implementation of processing on hierarchical data, and a method facilitating comprehension of the structure, meaning, and format of hierarchical data. There is also disclosed a data processing device including a user interface portion receiving data with tuples each a combination of data of different types, and lists each a series of data of a same type, display data obtained by converting the data, and an instruction requesting processing from a user, and presenting the display data, data managing portion managing the hierarchical data, display-data generating portion generating the display data from the hierarchical data, and analyzing portion implementing the processing. Data is presented in a table format having rows displaying level names and attribute names, respectively. The generating portion samples data of each level of hierarchical data to extract an amount possible to present, and converts the data format.

Abstract: An image retrieval system, including: an image input section for inputting a retrieval target image; an image data base for storing image data; a retrieval section for retrieving a plurality of images similar to said retrieval target image, by using a feature amount of said retrieval target image from said image data base; a filtering processing section for carrying out filtering processing using a whole image feature amount of another feature amount different from said feature amount of said retrieval target image about said plurality of images, to obtain a filtering result; and an outputting section for outputting said filtering result.

Abstract: An image retrieval system, including: an image input section for inputting a retrieval target image; an image data base for storing image data; a retrieval section for retrieving a plurality of images similar to said retrieval target image, by using a feature amount of said retrieval target image from said image data base; a filtering processing section for carrying out filtering processing using a whole image feature amount of another feature amount different from said feature amount of said retrieval target image about said plurality of images, to obtain a filtering result; and an outputting section for outputting said filtering result.

Abstract: In a similar face retrieval system for retrieving an image photographing a face similar to a face detected from a retrieval query image from a retrieval target image group by using an image photographing a human face as the retrieval query image, whole image features as features representative of background information are extracted from each whole area of an each image of a retrieval target image group, to calculate a degree of similarity through comparison with each set of whole image features, and an image having a degree of similarity not lower than a certain value and having a lower retrieval result order from retrieval results. It is possible to efficiently retrieve the same person playing in different scenes by utilizing different features for a retrieval process and a filtering process.

Abstract: For hierarchical data including tuples each including a combination of different data types and lists each listing data of a same data type, corresponding attribute-based data divided by attribute is stored in attribute-based files whereas information on the structure of the hierarchical data is held as schema information. The schema information includes data type information on the hierarchical data structure, including data type information on each element of each tuple and data type information on only a top element of each list. The attribute-based files storing the attribute-based data are managed to be in order by a file name management table. Data on each attribute is stored in a file in a state of being hierarchized in a list format corresponding to the depth at which the each attribute is listed in the schema information.

Abstract: To allow a user to use an appropriate search engine, this invention provides a computer system including: a client computer; a search engine; and a management computer, in which the management computer is configured to: collect logs of access; specify a parameter transferred from the client computer to an access destination of the client computer by analyzing the collected logs of access; judge that an access including a parameter judged as a search query is an access to the search engine; select a log of access to the search engine from the collected logs of access; extract an address of the search engine and the search query from the selected log of access to the search engine; and store a correspondence between the extracted address of the search engine and the extracted search query in a search engine profile.

Abstract: There is disclosed an interface enabling easy implementation of processing on hierarchical data, and a method facilitating comprehension of the structure, meaning, and format of hierarchical data. There is also disclosed a data processing device including a user interface portion receiving data with tuples each a combination of data of different types, and lists each a series of data of a same type, display data obtained by converting the data, and an instruction requesting processing from a user, and presenting the display data, data managing portion managing the hierarchical data, display-data generating portion generating the display data from the hierarchical data, and analyzing portion implementing the processing. Data is presented in a table format having rows displaying level names and attribute names, respectively. The generating portion samples data of each level of hierarchical data to extract an amount possible to present, and converts the data format.

Abstract: A data item name constituting a source data schema is extracted (1051), a data item name constituting a target data schema is extracted (1052), element concepts included in the extracted data item names of the source data schema and the target data schema are extracted (1053), a distinctive relation between the data item names is set based on the extracted element concepts to be stored in an ontology (1061), the data item names of the source data schema and the target data schema are read, a correlation is set between the two read data item names based on the distinctive relation to be stored in a mapping definition file (1071).

Abstract: For hierarchical data including tuples each including a combination of different data types and lists each listing data of a same data type, corresponding attribute-based data divided by attribute is stored in attribute-based files whereas information on the structure of the hierarchical data is held as schema information. The schema information includes data type information on the hierarchical data structure, including data type information on each element of each tuple and data type information on only a top element of each list. The attribute-based files storing the attribute-based data are managed to be in order by a file name management table. Data on each attribute is stored in a file in a state of being hierarchized in a list format corresponding to the depth at which the each attribute is listed in the schema information.

Abstract: In a similar face retrieval system for retrieving an image photographing a face similar to a face detected from a retrieval query image from a retrieval target image group by using an image photographing a human face as the retrieval query image, whole image features as features representative of background information are extracted from each whole area of an each image of a retrieval target image group, to calculate a degree of similarity through comparison with each set of whole image features, and an image having a degree of similarity not lower than a certain value and having a lower retrieval result order from retrieval results. It is possible to efficiently retrieve the same person playing in different scenes by utilizing different features for a retrieval process and a filtering process.

Abstract: To display many objects within a limited display region so a user can easily select the desired object, the invention includes: processing of arranging the objects in a virtual plane, processing of arranging a projection surface (a virtual projection body) on the virtual plane, processing of computing positions of the objects on the projection surface corresponding to the positions of the objects on the virtual plane, processing of converting the positions on the projection surface into positions on display regions of a display device, processing of displaying the objects on the display regions, and processing of operating the objects on the display regions based on instructions from a pointing device.

Abstract: A text classification apparatus directed to a plurality of languages, includes a unit for extracting information for converting a word from non-classified (unlabeled) texts, in a plurality of languages, into a word sense, a unit for learning a classification knowledge at a word sense level after converting a word extracted from a labeled text into a word sense, a unit for learning a classification knowledge at a word level from the labeled text, a unit for learning the classification knowledge at the word level from the classification knowledge at the word sense level and information on a relation between words extracted from the unlabeled text, and a unit for combining the respective classification knowledges to assign a category.

Abstract: A community is extracted by executing steps of: clustering relationship data; extracting a communication core of a relationship network; mapping the communication core to a dendrogram of relationship data; forming a community by using the dendrogram in accordance with a similarity degree of relationship data while the cluster is expanded; and aggregating communities. A community of a set of persons having high density relationships based on common topics and interests can be extracted from a set of human relationships and relationship data representative of the human relationships.

Abstract: To allow a user to use an appropriate search engine, this invention provides a computer system including: a client computer; a search engine; and a management computer, in which the management computer is configured to: collect logs of access; specify a parameter transferred from the client computer to an access destination of the client computer by analyzing the collected logs of access; judge that an access including a parameter judged as a search query is an access to the search engine; select a log of access to the search engine from the collected logs of access; extract an address of the search engine and the search query from the selected log of access to the search engine; and store a correspondence between the extracted address of the search engine and the extracted search query in a search engine profile.

Abstract: A personal-connection analysis and search method includes the steps of extracting a word from a text used in communications, accumulating the number of times of appearing the word as a communication flow, generating communication networks in an organization, extracting communication cores having a high density communication from the communication networks, calculating a communication ratio as a communication core formal degree between two constituent members having a ladder-connection relationship with respect to total communications in the communication core, and calculating a ratio of formal communication flow as a communication formal degree between the two constituent members with respect to total communication flows between the two constituent members, thereby quantifying a index for judging whether communications in an organization of the enterprise are strongly relevant to a business operation.

Abstract: A data item name constituting a source data schema is extracted (1051), a data item name constituting a target data schema is extracted (1052), element concepts included in the extracted data item names of the source data schema and the target data schema are extracted (1053), a distinctive relation between the data item names is set based on the extracted element concepts to be stored in an ontology (1061), the data item names of the source data schema and the target data schema are read, a correlation is set between the two read data item names based on the distinctive relation to be stored in a mapping definition file (1071).

Abstract: Provided is a program recommendation method for a computer system including: a program distribution server in which program information is stored, a schedule storing unit for storing a schedule of a user, and a program recommendation unit for recommending programs from the program information, comprising the steps of: recommending, in case of which a schedule to be an object of program recommendation is stored in the schedule storing unit, at least a program from the program information stored in the program distribution server based on the stored schedule; searching for a period that is before a start time of the schedule of an object of program recommendation and in which it is possible for the user to watch the programs recommended by the program recommendation unit; and registering a program watching schedule, which is a schedule for watching programs recommended in the searched period, in the schedule storing unit.

Abstract: In a user profiler used for program recommendation, a user profile which differs particularly according to the season is extracted, saved and appropriately re-used. For this purpose, there are provided a user profiler which extracts user profiles based on program information and a user's view history, and a profile memory which incrementally saves the profiles in a time series based on time information acquired from a timer in the user profiles. The profile memory generates and saves total profile information which is obtained by summing the profiles to be saved for predetermined periods. A total profile memory is also provided which saves the total profile information for these periods when a fixed time has elapsed, and the total profile information which was saved by the total profile memory is re-saved by the profile memory at a certain specified time.

Abstract: To display many objects within a limited display region so a user can easily select the desired object, the invention includes: processing of arranging the objects in a virtual plane, processing of arranging a projection surface (a virtual projection body) on the virtual plane, processing of computing positions of the objects on the projection surface corresponding to the positions of the objects on the virtual plane, processing of converting the positions on the projection surface into positions on display regions of a display device, processing of displaying the objects on the display regions, and processing of operating the objects on the display regions based on instructions from a pointing device.