JP2006139487A - Information summary presentation device - Google Patents

Abstract

Information is presented so that a user can easily grasp an overall image of an information group to be browsed.
Information including a search word is searched from a database by an information group search unit, and a feature word and class corresponding to this information are extracted from a feature word index. Each feature word is extracted by a browsing attribute classification unit. Each information is classified into browsing attributes based on the class to which it belongs, an information summary structure creation unit 7 creates a menu with a multifaceted expression indicating the viewing attributes on each side, and the information summary display unit 8 displays this menu.
[Selection] Figure 1

Description

  The present invention relates to an information summary presentation device that can present a summary of information searched by a user from a large amount of information group.

  Development of a technique for presenting information that has been narrowed down to some extent from an information group in which a large amount of various attributes (field, topic, medium, author, date, viewing purpose, etc.) are mixed is underway (for example, Patent Document 1, 2).

  For example, when narrowing down by keyword search on the Internet, depending on the specificity of the keyword, the more general the keyword is, the more the documents including various attributes are presented in the search result.

  As such an information presentation apparatus, there are a list type information presentation apparatus, a related display type information presentation apparatus, a classification type information presentation apparatus, and an information presentation apparatus using distortion.

  The list type information presentation device presents a part of a list summarizing an information group or a large amount of information as a search result to a user.

  The related display type information presentation device is a device that arranges and presents information summary words, titles, and the like in a characteristic space according to a distance based on the document similarity of the information. The relevance includes a semantic relevance and a structural relevance such as a hyperlink and a network device configuration. Some presentations express structural relationships by connecting them with lines.

  The classification type information presentation apparatus narrows down and expands information viewpoints by classifying information manually, semi-automatically, or automatically from a single viewpoint or a combined viewpoint. This classification result is presented in a hierarchical tree structure.

When presenting a lot of information and classification information in a certain space, an information presentation device that uses distortion expands only those that are closely related to the point of interest and those that are less relevant. To display. This space is, for example, a rectangle, a circle, a sphere, etc., and the information to be watched in the space is large, and the one with low relevance is displayed in a size corresponding to the degree of relevance. It presents the entire information in a limited space.
JP-A-9-16627 JP 2000-293537 A

  However, the list-type information presentation device needs to see a large number of mixed information groups and summaries when selecting information presented by the user, and it takes a lot of trouble and grasps the entire information group. There is a problem that can not be. In addition, the user may stop searching for information, and may not be able to reach the information that is being searched for.

  The related display type information presentation device, the classification type information presentation device, and the information presentation device using distortion are compressed and presented when the search results are presented with a large number of words and document titles. It is difficult to grasp the whole picture of the information group.

  As described above, the conventional information presentation apparatus has no problem when the number of words or document titles included in the target information group is small. However, when the number of these information increases, it is possible to grasp the balance between the number and attributes. It was difficult. Also, when classifying and presenting search results, if many classification classes appear, it becomes difficult to present words, document titles, etc. in a limited space, and it is impossible to grasp the entire summary of the information group There was a problem.

  Further, when presenting a large amount of information summaries, conventional information presentation devices present them as a list or present them hierarchically. For this reason, there is a problem that the information located behind the enumeration and the information located below the hierarchy have a lot of trouble to select, and as a result, are not used much.

  The present invention has been made in view of the above, and an object of the present invention is to make it possible to present information that allows the user to easily grasp the entire image of the information group to be browsed.

  In order to solve the above problems, an information summary presentation device according to the present invention includes information group collection means for collecting information to be searched in advance and storing it in a database, a feature word for each information, and a class to which the feature word belongs And classifying feature expression extracting means for associating information, feature words and classes with each other and storing them in a database as a feature word index, search word input means for accepting input of a search word, and information including the search word Information group search means for extracting a feature word corresponding to this information and a class to which the feature word belongs from a feature word index, and browsing each information based on the class to which each extracted feature word belongs Multi-faceted expression that includes browsing attribute classification means for classifying into attributes and the number of classified browsing attributes, and showing the browsing attributes on each side Information general structure creation means for creating a menu with, and having a information summary display means for displaying said menu.

  In the present invention, information including a search word is searched, and feature words and classes corresponding to this information are extracted, and each information is classified into browsing attributes based on the class to which each feature word belongs. By creating and displaying a menu with multi-faceted expressions indicating browsing attributes, the user who viewed this menu can grasp the outline of information to be browsed as the types and number of browsing attributes.

  According to the present invention, the user can easily grasp the entire image of the information group to be browsed.

  Hereinafter, an embodiment will be described.

  FIG. 1 is a functional block diagram showing the configuration of the information summary presentation device in the present embodiment. The system shown in FIG. 1 includes a database 1, an information group collection unit 2, a class-specific feature expression extraction unit 3, a search word input unit 4, an information group search unit 5, a browsing attribute classification unit 6, an information summary structure creation unit 7, an information summary. A display unit 8 is provided. This information summary presentation device is configured by a computer, and the processing of each unit is executed by a program.

  The database 1 stores an information group collected by the information group collection unit 2 and also stores a feature word index in which feature words and classes are associated with each piece of information included in the information group.

  The information group collection unit 2 collects information provided by various servers on the Internet in advance and stores the information in the database 1.

  The class-specific feature expression extraction unit 3 extracts a feature word for each piece of information, extracts a class to which the feature word belongs, and stores the information, the feature word, and the class in the database 1 as a feature word index in association with each other. For example, proper nouns are targeted as feature words.

  The class to which the feature word belongs is unique including at least one of person name, occupation name, disease name, color name, organization name, place name, facility name, product name, event name, natural object name, title name, and unit name Numeric value including at least one of expression, time expression including at least one of time and period, amount, index, points, ratio, multiple, frequency, rank, age, latitude / longitude, size, number, ordinal number As a representation, the class-specific feature representation extraction unit 3 extracts at least one of these unique representation, temporal representation, and numerical representation.

  When a search keyword related to information that the user 10 wants to know is input using the search word input unit 4, the search word input unit 4 receives the search keyword and passes it to the information group search unit 5. The search word input unit 4 may accept a search keyword input from a keyboard or the like connected to the apparatus, or accept a search keyword input from a terminal connected to the apparatus via a network. It may be a thing.

  The information group search unit 5 searches the information group stored in the database 1 for information including the search keyword and sets it as an information group to be presented. The information group search unit 5 extracts not only the information including the search keyword but also the feature words included therein and the class to which the feature word belongs from the feature word index, and passes them all to the browsing attribute classification unit 6.

  The browsing attribute classification unit 6 classifies each information into browsing attributes based on the class to which each extracted feature word belongs. For this classification, for example, the degree to which each class belongs to each browsing attribute is obtained in advance by a probability model, and correspondence table information in which each class is associated with the probability that each class belongs to each browsing attribute is stored in the database 1. The browsing attribute classification unit 6 classifies each class as a browsing attribute with the highest probability based on the correspondence table, using the feature expression vector with class of each information as an element.

  Viewing attributes include fields (politics, economics, science, society, region, sports, performing arts, etc.), topic words, outlines (specs, summaries, etc.), media (EC, Web news, TV program guide) Product description page, BLOG page, etc.), author, date and time, evaluation (recommended comment, rank, etc.), browsing purpose, and at least one of the browsing purposes. Store in database 1.

  The information outline structure creating unit 7 creates a menu with multi-faceted expressions including the number of classified browsing attributes and showing the viewing attributes on each surface. This menu is, for example, as shown in FIG. FIG. 2A shows an example in which “case, region, trial, person name” is shown as a viewing attribute, and “starring actor, movie theater, DVD, story” is shown in FIG. 2B. In any case, the number of browsing attributes is 4, and the browsing attributes are two-dimensionally arranged.

  The information summary display unit 8 displays this menu on the screen. The information summary display unit 8 may cause the menu to be displayed on the screen of the information summary presentation device or may be displayed on the screen of a user terminal connected via a network. Here, when the user clicks the browsing attribute on the menu, the information outline structure creating unit 7 creates a menu so that a search result of information related to the viewing attribute is displayed.

  FIG. 3 is a diagram illustrating another example when the browsing attributes are two-dimensionally arranged for the menu to be presented. The larger the number of browsing attributes (dimensions), the larger the number of surfaces indicating the browsing attributes, and the fewer the dimensions, the smaller the number of surfaces indicating the browsing attributes. However, each menu shows a state in which a plane indicating the viewing attribute is arranged at an equal distance from the center of the display.

  FIG. 4 is a diagram illustrating an example in which browsing attributes are three-dimensionally arranged for a menu to be presented. Also here, a state is shown in which planes indicating viewing attributes are arranged at equal distances from the center of the display.

  Therefore, according to the present embodiment, the information group search unit 5 searches the database 1 for information including the search keyword, extracts a feature word and class corresponding to this information from the feature word index, and browse attribute classification unit 6 classifies each information into browsing attributes based on the class to which each feature word belongs, creates a multi-faceted expression menu indicating browsing attributes on each side by the information summary structure creation unit 7, and displays this information by the information summary display unit 8. By displaying the menu, the user who sees this menu can grasp the outline of the information group to be browsed as the type and number of browsing attributes, so it is easy to grasp the whole picture of the information group. be able to.

  Further, according to the present embodiment, when a viewing attribute on the menu is clicked after giving a general effect to the information group to be presented by menu display, information related to the viewing attribute is displayed. By displaying this search result, the trouble of browsing subsequent partial information becomes constant, there is no information that is difficult to use, and all information can be browsed in the same manner.

It is a functional block diagram which shows the structure of the information summary presentation apparatus in one embodiment. (A) in the figure is an example of a menu indicating “case, region, trial, personal name” as a viewing attribute, and (b) in FIG. 10 is a menu indicating “leading actor, movie theater, DVD, story” as the browsing attribute. It is a figure which shows the example of. It is a figure which shows the example at the time of arrange | positioning the browsing attribute about a menu two-dimensionally. It is a figure which shows the example at the time of arrange | positioning the browsing attribute about a menu three-dimensionally.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Database, 2 ... Information group collection part 3 ... Feature expression extraction part, 4 ... Search word input part 5 ... Information group search part, 6 ... Browsing attribute classification part 7 ... Information summary structure creation part, 8 ... Information summary display part 10 ... User

Claims (5)

Information group collecting means for collecting information to be searched in advance and storing it in a database;
A feature word extraction means for extracting a feature word and a class to which the feature word belongs for each information, and storing the information, the feature word, and the class in a database as a feature word index in association with each other;
Search word input means for accepting input of a search word;
Information group search means for searching the information including the search word from the database, and extracting from the feature word index a feature word corresponding to this information and a class to which the feature word belongs,
Browsing attribute classification means for classifying each information into browsing attributes based on the class to which each extracted feature word belongs;
An information summary structure creation means for creating a menu with a multifaceted expression that includes the number of classified browsing attributes, and showing the browsing attributes on each side;
Information summary display means for displaying the menu;
An information summary presentation device characterized by comprising:   The class-specific feature expression extracting means includes at least one of a person name, occupation name, disease name, color name, organization name, place name, facility name, product name, event name, natural object name, title name, and unit name as a class. A unique expression including one, a time expression including at least one of time and period, and an amount, index, point, ratio, multiple, frequency, rank, age, latitude / longitude, size, number, ordinal number The information summary presentation device according to claim 1, wherein at least one of numerical expressions including at least one is extracted. The correspondence table information that associates each class with the probability that each class belongs to each browsing attribute is stored in the database,
3. The information summary presenting apparatus according to claim 1, wherein the browsing attribute classification means classifies each class into a browsing attribute with the highest probability based on the correspondence table.   The information summary presentation device according to any one of claims 1 to 3, wherein the information summary structure creation unit creates a menu in which each surface indicating a viewing attribute is arranged at an equal distance from a display center. 5. The information summary presentation device according to claim 1, wherein the browsing attribute includes at least one of a field to which the information belongs, a topic word, a media medium, an author, a date, and an evaluation. .

Images