JP2008139942A - Information processing apparatus, information processing method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information processing apparatus, an information processing method and a program that can significantly reduce a load of digital content management. <P>SOLUTION: The information processing apparatus 1 refers to the category of content to be processed, acquires content related to the category, extracts the most significant word from the acquired content, and includes the extracted most significant word in metadata on the target content. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

The present invention relates to an information processing apparatus, an information processing method for processing content such as a still image and a moving image,
And related to the program.

In recent years, content such as images stored by individuals has been increasing. Digital image data, which is representative content, is often added with metadata including information indicating the shooting date and time by the function of the digital camera. By using the information included in this metadata, for example, digital image data captured on a specific day can be searched, which helps to manage the digital image data. However, information included in general metadata is information indicating the shooting date and time and shooting conditions (ISO sensitivity, shutter speed, etc.), and does not include information related to the content. For this reason, it is difficult to perform management based on the contents even if information included in the metadata is used. Thus, conventionally, an apparatus having a function of searching for an event related to image data has been proposed (see, for example, Patent Document 1).
. This apparatus stores information related to the date and time of the event, and determines whether or not the image data and the event are related based on the shooting date and time of the target image data.
JP 2003-303210 A

However, the conventional apparatus needs to prepare accurate information regarding the date and time of the event. For this reason, the user must always input the latest accurate information, and there is a problem that the load is large.
Further, in order to efficiently manage the content, the user himself / herself may add information related to the content content in the form of metadata, for example. However, manually inputting data for each piece of content is very complicated and heavy, and a large number of pieces of content cannot be processed. Therefore, a method for easily and efficiently managing a large number of contents has been demanded.
An object of the present invention is to provide an information processing apparatus, an information processing method, and a program that can efficiently manage a large number of contents and can significantly reduce the load related to the management of contents.

In order to achieve the above object, the present invention acquires attribute information related to the contents of the processing target content, acquires reference content related to the attribute information, and the reference content acquisition unit acquires the reference content An information processing apparatus comprising: extraction means for extracting feature information from reference content; and additional information generation means for generating additional information of the processing target content including the feature information extracted by the extraction means. provide.
According to this configuration, the feature information is extracted from the reference content acquired based on the attribute information related to the content of the processing target content, and the additional information of the processing target content including the feature information is generated. Can be included in the additional information. If the additional information generated here is used as, for example, a search key, it becomes possible to search the processing target content based on the content, and the content can be managed efficiently. Further, when generating the additional information, reference content related to the attribute information related to the content of the processing target content is acquired. Therefore, if there is attribute information corresponding to the processing target content, the additional information is automatically generated. . For this reason, it is not necessary for the user to perform complicated work regarding generation of additional information, and the load related to content management can be greatly reduced.

Here, a standard expression storage unit that stores a standard expression for each attribute information, and a standard expression acquisition unit that acquires a standard expression corresponding to the attribute information of the processing target content from the standard expression storage unit, The reference content acquisition unit may be configured to acquire the reference content including a reference expression acquired by the reference expression acquisition unit or an expression similar to the acquired reference expression.
In this case, since the reference content includes the standard expression corresponding to the attribute information or an expression similar to the standard expression, the reference content unrelated to the content of the processing target content is not acquired. For this reason, the feature information extracted from the reference content is certainly information related to the content of the processing target content, and it can be said that the reliability of the feature information is very high. Therefore,
The additional information of the processing target content includes information having high relevance to the content, and the content can be managed extremely efficiently.

In addition, when the plurality of reference contents are acquired by the reference content acquisition unit, the extraction unit extracts candidate feature information from the plurality of reference contents, and the appearance state of the reference expression Content evaluation means for weighting each reference content based on the feature information, and extracting the feature information from a plurality of feature information candidates based on the weight of the reference content from which the feature information candidates are extracted And feature information extraction means.
In this case, the reference content is weighted based on the appearance state of the standard expression, and further, the feature based on the weight of the reference content that is the origin from among the candidate feature information extracted from the reference content. Information is selected. For this reason, the feature information is extracted from the reference content having high relevance to the content of the processing target content among the feature information candidates. Therefore, the additional information of the processing target content includes information having high relevance with the content, and the content can be managed extremely efficiently.

Further, the candidate extraction unit divides the language expression included in the reference content into language units, performs weighting on each language unit, and based on the weights, part of the language expressions is performed in the language unit. It may be selected as a candidate for the feature information.
In this case, when extracting feature information candidates, weighting is performed after the language expression is divided in units of languages, so that language expressions having high importance in the reference content can be extracted as feature information candidates. For this reason, the feature information candidates are highly relevant to the content of the processing target content. Furthermore, since the feature information candidates are linguistic expressions, it is very easy for the user to understand and can be used easily and effectively as a search key. Therefore, the additional information of the processing target content is easy to recognize and the searchability is high, so that the content can be managed very efficiently.

Further, the reference expression may be a linguistic expression composed of any language unit of a word, a phrase, a clause, a sentence, and a paragraph.
In this case, since the reference expression is a linguistic expression composed of any language unit of words, phrases, sections, sentences, and paragraphs, the user can easily determine whether or not the reference expression is appropriate, The standard expression can be easily optimized. As a result, it is possible to acquire reference content having a high relevance to the content of the processing target content, and as a result, the reliability of the feature information can be further increased, and the content can be managed more efficiently.

In addition, an input unit for inputting the attribute information may be provided, and the reference content acquisition unit may acquire the reference content related to the attribute information input by the input unit.
In this case, the reference content is acquired by inputting the attribute information of the processing target content, and additional information of the processing target content is generated. As a result, useful additional information is generated simply by performing an operation of inputting additional information, so that the load related to content management can be greatly reduced.

Furthermore, the reference content acquisition means may acquire the reference content created or updated within a predetermined time from the date and time when the processing target content was created or updated.
In this case, since the reference content is created or updated at a time close to the date and time when the processing target content was created or updated, the information included in the reference content is surely related to the details of the processing target content. It will have a sex. In other words, the reference content deviating from the content of the processing target content can be excluded, and as a result, the reliability of the feature information can be further improved, and the content can be managed more efficiently.

Furthermore, it is good also as a structure provided with the communication means which transmits / receives information via the communication line connected externally, and the said reference content acquisition means acquires the said reference content from the external apparatus communicable by the said communication means.
In this case, by acquiring reference content via an externally connected communication line, content suitable as reference content can be acquired from an extremely large number of content.

In addition, an external content display unit that displays the content received by the communication unit may be provided, and the reference content acquisition unit may acquire the content displayed in the past by the external content display unit as the reference content.
In this case, since the reference content can be acquired from the displayed content in an operation other than the process of generating the additional information of the processing target content, more preferable reference content can be acquired.

The present invention acquires attribute information relating to the content of the processing target content, acquires reference content related to the attribute information, extracts feature information from the acquired reference content, and includes the extracted feature information An information processing method characterized by generating additional information is provided.
According to this method, the feature information is extracted from the reference content acquired based on the attribute information regarding the content of the processing target content, and the additional information of the processing target content including the feature information is generated. Can be included in the additional information. The additional information generated here can be used when performing a search based on the content, and the content can be managed efficiently. In addition, since additional information is generated if there is attribute information related to the content of the processing target content, it is not necessary for the user to perform complicated work for generating the additional information, and the load related to content management can be greatly reduced.

The present invention also provides a computer for acquiring attribute information related to the content of the processing target content, acquiring reference content related to the attribute information, and the reference content acquired by the reference content acquiring means. There is provided a program for functioning as extraction means for extracting feature information from, and additional information generation means for generating additional information of the processing target content including the feature information extracted by the extraction means.
By executing this program by the computer, the computer extracts feature information from the reference content acquired based on the attribute information regarding the content of the processing target content, and generates additional information of the processing target content including the feature information. To do. For this reason, information related to the content of the processing target content can be included in the additional information. The additional information generated here can be used when performing a search based on the content, and the content can be managed efficiently. In addition, if there is attribute information related to the content of the processing target content, the computer generates additional information, so that the user does not have to perform complicated operations for generating additional information, and the load related to content management can be greatly reduced. .

Embodiments to which the present invention is applied will be described below with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of an information processing apparatus 1 according to an embodiment to which the present invention is applied.
The information processing apparatus 1 selects a word related to the content of digital content and executes metadata generation processing for generating metadata (additional information) including the selected word.
In this metadata generation processing, the information processing apparatus 1 uses other contents (
Hereinafter, the registered content created or updated at a date close to the creation date of the digital content to be processed (hereinafter referred to as target content) is selected and acquired. In the information processing apparatus 1, a field representing the content of the target content is designated by a user input operation. The information processing apparatus 1 selects the acquired registered content based on the keyword corresponding to the input target content field, and extracts an important word from each selected registered content. Further, the information processing apparatus 1 selects the most important word (feature information) from a plurality of important words (feature information candidates) extracted from a plurality of registered contents, and generates metadata including the most important word. Thus, the metadata of the target content is used.
Here, the target content includes all so-called content such as image data of moving images and still images, audio data, and the like, and there is no restriction on the recording method and the like, and even analog content is digital content. May be. In the present embodiment, as an example, a case where digital content including digital image data and the like is processed will be described.
First, the configuration of the information processing apparatus 1 will be described.

As shown in FIG. 1, the information processing apparatus 1 includes a control unit 2 that executes control of each unit of the information processing apparatus 1 and processing that will be described later, and a ROM 3 that stores various control programs executed by the control unit 2. And a RAM 4 for temporarily storing programs and data.
In addition, the information processing apparatus 1 includes a storage unit 5 that stores a program executed by the control unit 2 and data processed when the program is executed, and an instruction input by the user to the information processing apparatus 1. An input unit 6, an output unit 7 for outputting the results of processing executed by the control unit 2, and an external interface (I / F) 8 connected to an external device (not shown)
And a communication interface 9 connected to an external communication line 15. These units are connected to each other by a bus 10.

The storage unit 5 stores various information in a nonvolatile manner using a magnetic recording medium, an optical recording medium, or a semiconductor storage device. The storage unit 5 stores an application program 51 for realizing various functions of the information processing apparatus 1. In addition, the storage unit 5 stores a metadata generation program 52 for executing metadata generation processing, which will be described later with reference to FIGS. 4 to 6, in the information processing apparatus 1.
Further, the storage unit 5 has a plurality of data storage areas, and in these data storage areas, an application data storage unit 53 for storing data generated when the application program 51 is executed, a field for storing keywords by field. Another keyword information storage unit 54, a target content storage unit 55 that stores target content, a registered content information storage unit 56 that stores information related to registered content, and an important word storage unit 57 that stores information generated in the metadata generation process And the most important word storage unit 58 is included.

The application program 51 is a program for realizing various functions including a web browser function and an e-mail transmission / reception function by the information processing apparatus 1. The web browser function connects to a network 20 such as the Internet via an external communication line 15 connected to the communication interface 9, accesses a server (not shown) via the network 20, and retrieves web content on the server. This is a function for receiving and displaying on the display screen 71. When this web browser function is executed, the control unit 2 functions as external content display means. Web content is content composed of one or more files including text data and image data described in a format such as HTML. There are no restrictions on the format of various files constituting the web content, and links to other files may be included.

When executing the web browser function, the information processing apparatus 1 stores the displayed web content history in the application data storage unit 53 as a browser history. This display history is, for example, a URL (Uniform Resource Locat) for acquiring the displayed web content.
or) and the displayed date and time.
The e-mail transmission / reception function is connected to the network 20 via the communication line 15 connected to the communication interface 9, transmits / receives e-mail via the network 20, and sends the transmitted / received e-mail to the application data storage unit 53. Remember. The e-mail transmission / reception function may be realized as a so-called web mail function. In this case, the function of actually transmitting / receiving electronic mail is executed by a server (not shown) on the network 20, and the transmitted / received electronic mail is stored in the server. The information processing apparatus 1 sends and receives e-mails by accessing the server using a predetermined ID and password,
You can get sent and received emails.
The application data storage unit 53 is an area for storing data generated or acquired when the application program 51 is executed. As described above, the application data storage unit 53 stores a browser history, transmitted and received e-mails, and the like.

The field-specific keyword information storage unit 54 (standard expression storage means) is an area for storing keywords (standard expressions) for each field for a plurality of fields (attribute information).
FIG. 2 is a diagram schematically illustrating an example of keywords stored in the field-specific keyword information storage unit 54.
As shown in FIG. 2, examples of the field include overseas travel, domestic travel, school events, circles, overseas business trips, domestic business trips, family events, and the like. In addition, in connection with each field, connection, waiting time, direct flight, voyage, train, hotel, athletic meet, school bag, photo session, hotel,
Keywords related to each field such as meetings and weddings are stored.
These fields and keywords corresponding to each field can be freely input by the user by operating the input unit 6, and are not limited to the example of FIG. 2, and are suitable for target content processed by the information processing apparatus 1. Preferably, the field and keyword are stored. That is,
The field illustrated in FIG. 2 is suitable for, for example, a still image or a moving image captured by the user, but when other types of target content are used, the field is not limited to the above example. For example, even if information such as sports, drama, variety, animation, action, violence, suspense, musical, adult, horror, family, children, music, politics, education, etc. is used as the field Good.
FIG. 2 shows an example in which words are stored as keywords, but the keywords are not limited to words, and language expressions composed of language units such as phrases, clauses, sentences, and paragraphs can be used.

The target content storage unit 55 is an area for storing content that is a target of metadata generation processing executed by the information processing apparatus 1. The target content storage unit 55 stores the metadata of the target content in addition to the data of the target content itself. This metadata is Ex
It may be added to the data of the target content itself in a standardized format such as if (Exchangeable Image File Format) or may be stored in the target content storage unit 55 as data independent of the target content.
The target content stored in the target content storage unit 55 is, for example, an external hard disk drive or a storage device incorporating a semiconductor memory, a personal computer, a digital still camera, a digital video camera, a mobile phone via the external interface 8 or the like. It is input to the information processing apparatus 1 from a device such as a type telephone. When the information processing apparatus 1 itself includes an imaging unit that captures a still image or a moving image, the image data captured by the imaging unit may be stored in the target content storage unit 55 as target content.

The registered content information storage unit 56 is an area for storing information for acquiring registered content in the metadata generation process.
FIG. 3 is a diagram illustrating a configuration example of information stored in the registered content information storage unit 56.
As shown in FIG. 3, the registered content information storage unit 56 stores content names and content acquisition information for acquiring each content in association with each other. Examples of the registered content include web content (web page, blog format web page, still image data, moving image data, audio data, etc.) that can be acquired via the network 20 such as the Internet, and e-mail. For information that can be acquired via the network 20, for example, a URL is stored as content acquisition information. The content acquisition information corresponding to the e-mail is a path for referring to the transmitted / received e-mail data stored in the application data storage unit 53, for example.

The important word storage unit 57 is an area for temporarily storing the important words extracted in the metadata generation process, and the most important word storage unit 58 is an area for storing the most important words extracted in the metadata generation process. It is.
The input unit 6 includes various key switches such as keys for inputting numbers, letters, symbols, etc., function keys for instructing a specific operation, a rotary switch, a power ON / OFF button, and the like. A corresponding operation signal is generated and output to the control unit 2.
The output unit 7 outputs a processing result by the control unit 2 and an image being processed. Specifically, LC
It has a function of displaying an image on a display screen 71 constituted by D (liquid crystal display) or the like, or a function of printing an image by a printing device (not shown).
The information processing apparatus 1 uses the input unit 6 and the output unit 7 to use the field-specific keyword information storage unit 5.
It is possible to accept input of fields and keywords stored in 4. Specifically, the information processing apparatus 1 displays a keyword input screen on the display screen 71 under the control of the control unit 2, and operates the input unit 6 while the keyword input screen is displayed. The input character string is acquired as a field and a keyword and stored in the field-specific keyword information storage unit 54.

The external interface 8 is an interface connected to a device external to the information processing device 1. Here, examples of the external device include a storage device incorporating a hard disk drive and a semiconductor memory, a personal computer, a digital still camera, a digital video camera, and a mobile phone. External interface 8 is USB (Universal Serial
Bus), IEEE1394, Bluetooth (registered trademark), UWB, ZigBee (registered trademark) and other connectors that correspond to device connection methods, etc., and an interface circuit that can execute protocols corresponding to the various connection methods.
The communication interface 9 is connected to a communication line 15 outside the information processing apparatus 1, and transmits and receives various data using the network 20 such as the Internet via the communication line 15. Specific configurations of the communication interface 9 include a wired connection interface corresponding to a communication method such as Ethernet (registered trademark), a wireless communication interface corresponding to a communication method such as IEEE 802.11, and the like. The communication interface 9 includes a connector conforming to the above standard and a wireless communication unit, and further includes an interface circuit capable of executing a protocol conforming to the above standard.

The communication line 15 and the network 20 are configured to include various communication lines such as a dedicated line, a public telephone line, a satellite communication line, and a mobile communication line, various server devices (not shown), and the like. It is a communication line network, and its specific and physical aspects are not particularly limited. Various devices such as a server, a firewall, and a gateway device provided by a telecommunications carrier may be interposed between the devices connected to the network 20, but they are not shown here.

Next, the operation of the information processing apparatus 1 will be described. In the operation described below, the control unit 2 functions as a reference content acquisition unit, an extraction unit, an additional information generation unit, a reference expression acquisition unit, a candidate extraction unit, a content evaluation unit, and a feature information extraction unit.
FIG. 4 is a flowchart showing metadata generation processing executed by the information processing apparatus 1.
Prior to starting the metadata generation process shown in FIG. 4, the target content is stored in the target content storage unit 55 of the information processing apparatus 1.
Then, the control unit 2 controls the output unit 7 to guide the input of the target content field on the display screen 71 and acquire the target content field input by the operation of the input unit 6 (step S11). .

Subsequently, the control unit 2 executes an important word extraction process for the registered content based on the information stored in the registered content information storage unit 56 and the keywords stored in the field-specific keyword information storage unit 54. (Step S12). The process of step S12 will be described later with reference to FIG.
Furthermore, the control unit 2 uses the browser history stored in the application data storage unit 53 and the keywords stored in the field-specific keyword information storage unit 54 to determine the importance of information displayed by the web browser function. A word extraction process is executed (step S13).
). The process of step S13 will be described later with reference to FIG. 5 in the same manner as the process of step S12.

Thereafter, the control unit 2 executes the most important word extraction process based on the important words extracted by the important word extraction process in steps S12 and S13, and extracts one or a predetermined number of the most important words ( Step S14). The process of step S14 will be described later with reference to FIG.
And the control part 2 produces | generates the metadata containing the most important word extracted by step S14, and memorize | stores it in the object content memory | storage part 55 as metadata of object content (step S15), and complete | finishes this process. If the metadata of the target content has already been stored in the target content storage unit 55 in step S15, the control unit 2 determines that the target content storage unit 5
5 is updated by including the most important word in the metadata stored in 5.

FIG. 5 is a flowchart showing important word extraction processing.
As described above, in step S12 of FIG. 4, the important word extraction process described below is executed for the registered content. In step S13, an important word extraction process is executed for web contents displayed in the past by the web browser function.

In this important word extraction process, the control unit 2 acquires a keyword stored in the field-specific keyword information storage unit 54 in association with the field input in step S11 (FIG. 4) (step S21).
Subsequently, the control unit 2 refers to the metadata of the target content stored in the target content storage unit 55, and acquires the date and time when the target content was created or last updated (step S22). And the control part 2 acquires the information of the registration content memorize | stored in the registration content information memory | storage part 56, and the browser log | history memorize | stored in the data storage part 53 for applications, Out of the content acquired by these information , Select one content created or updated within a predetermined time from the date and time when the target content was created or updated (
Step S23). The content selected in step S23 is registered content or web content created or updated within a predetermined time regardless of whether the target content is created or updated before or after.
Thereafter, the control unit 2 acquires data of the selected content (step S24). Here, when the selected content is web content, the control unit 2 accesses the network 20 via the communication interface 9 and acquires the web content. When the selected content is stored in the application data storage unit 53, the control unit 2
Acquires content from the application data storage unit 53.

The control unit 2 determines whether the content acquired in step S24 includes any of the keywords acquired in step S21 or a similar word of the keyword (step S2).
5).
Here, the similar term is a synonym or synonym of a keyword, or a word that looks similar to a keyword.
When using synonyms or synonyms, for example, a similar word dictionary may be stored in the storage unit 5 in advance, and similar words may be selected from the similar word dictionary when a keyword is acquired in step S21. Further, when using a word that looks similar to a keyword, for example, step S25.
When searching for a keyword in the content, a search condition may be set as appropriate, and a word that matches only a part of the character with the keyword may be searched together with the keyword.
If the content includes a keyword or similar word, the control unit 2 determines this content as the important word extraction target content (step S26), and proceeds to step S27. If the content contains neither a keyword nor a similar word, the control unit 2 proceeds to step S27 as it is.

In step S27, in addition to the confirmed content, the control unit 2 stores the registered content information stored in the registered content information storage unit 56, and the application data storage unit 5
3, it is determined whether there is content that can be acquired from the browser history stored in 3. If there is another content, the control unit 2 returns to step S23, selects another content, and repeats the above processing.
As a result, among the contents created or updated within a predetermined time from the date and time when the target content is created or updated, the content including the keyword or similar word is determined as the important word extraction target content.

When the processing of steps S23 to S26 is completed for all the contents (step S27; No), the control unit 2 performs TF / IDF (Term Frequency & IDF) for each important word extraction target content determined in step S26. Inverse Document Frequency
) The process is executed, and important words are extracted from each important word extraction target content (step S28).

The process of step S28 will be described in detail.
The TF / IDF process is a process for calculating weights for all words included as text data in the important word extraction target content.
Text data of one important word extraction target content is d, and one of the words is t
In this case, the weight W (t, d) of the word t in the text data d is obtained by the following equation (1).
W (t, d) = TF (t, d) × IDF (t) (1)
Here, TF (t, d) is the appearance frequency of the word t in the text data d.
IDF (t) is calculated from the number of text data of the important word extraction target content in which the word t appears among all the important word extraction target content as the appearance frequency of the word t.

TF (t, d) is an index representing the importance of a word based on the appearance frequency in one important word extraction target content. In other words, words that frequently appear in the content are regarded as important words.
IDF (t) is an index indicating whether the word t is a word that appears specifically in a specific important word extraction target content. In other words, words that frequently appear in specific content are regarded as important words.
In step S28, the control unit 2 obtains weights W (t, d) for all the words included in the important word extraction target content by the calculation shown in the above formula (1), and indicates the weights W (t, d). A word with a large weight is extracted as an important word. This extraction is performed for all important word extraction target contents for each important word extraction target content. The number of important words to be extracted is not particularly limited, and the control unit 2 may extract a predetermined number of words, or all the words having W (t, d) exceeding a certain value. It may be extracted.

In subsequent step S29, the control unit 2 stores the important word extracted in step S28 in the important word storage unit 57 in association with the important word extraction target content from which the important word has been extracted.
Further, the control unit 2 proceeds to step S30, and for each important word extraction target content, the weight W (t, d) of the important word extracted from the important word extraction target content, appears in the important word extraction target content. The processing for obtaining the weight for each important word extraction target content is performed in consideration of the types of the keywords and similar words and the number of appearing keywords and similar words.
After executing the important word extraction process shown in FIG. 5 for both the registered content and the content obtainable from the browser history, the control unit 2 executes the most important word extraction process described below.

FIG. 6 is a flowchart showing the most important word extraction process.
First, the control unit 2 selects one of the important word extraction target contents determined by the important word extraction process shown in FIG. 5, and extracts the important words extracted from the important word extraction target contents and each important word. The word weight is read and acquired from the important word storage unit 57 (step S41).
).
Subsequently, the control unit 2 selects, for the selected important word extraction target content, the types of keywords and similar words that appear in the important word extraction target content, the number of appearing keywords and similar words, and the important words. The weight obtained in step S30 in FIG. 5 is acquired for the content to be extracted (step S42).

Then, based on the keyword and similar word type acquired in step S42, the number of appearing keywords and similar words, and the weight of the important word extraction target content,
The weight of the important word extracted from the important word extraction target content is recalculated (step S4).
3) The important word and the weight are stored again in the important word storage unit 57 (step S44).
The control unit 2 determines whether there is another important word extraction target content (step S45).
) If there is another important word extraction target content, the process returns to step S41, and another important word extraction target content is selected, and the above processing is repeated.
When the processing of steps S41 to S44 is completed for all important word extraction target contents (step S45; Yes), the control unit 2 selects each important word from the important words stored in the important word storage unit 57. The most important word is selected on the basis of the weight of and stored in the most important word storage unit 58 (step S46). Here, the number of the most important words selected by the control unit 2 is not particularly limited, and may be one or a plurality of the most important words, and may be limited with respect to the metadata capacity of the target content. Can be determined on the basis of.

As described above, when the user specifies a field related to the target content, the information processing apparatus 1 according to the embodiment to which the present invention is applied selects and selects the important word extraction target content related to the specified field. The most important word is selected from the content, and this most important word is registered in the metadata of the target content. Thereby, an appropriate word related to the content of the target content can be included in the metadata of the target content.
This metadata can be used when a desired content is searched from a large number of contents including the target content. In particular, if metadata including words related to the content content is used, it is possible to perform a search using a complicated condition, so that there is an advantage that the load required for content management can be greatly reduced. Furthermore, the information processing apparatus 1 performs the above-described processing only by the user specifying the field of the target content. In other words, the user only has to input the field, and compared with the case where the user manually creates metadata related to the content, the operation is extremely simple and the load on the user is greatly increased. Can be reduced.

Further, the information processing apparatus 1 stores keywords belonging to a plurality of fields in the field-specific keyword information storage unit 54 in advance, and stores keywords or similar keywords stored corresponding to the field of the target content designated by the user. Important word extraction target content is selected based on whether or not it is included. Thereby, an important word can be extracted by selecting an important word extraction target content that is reliably related to the content of the target content and extracting a word from these important word extraction target content.
Further, when extracting the important word, the information processing apparatus 1 performs weighting after dividing the text of the important word extraction target content into words, so that the important word extraction target content is surely a highly important word. Can be extracted as an important word.
Furthermore, since the information processing apparatus 1 extracts the most important word based on the weight of the important word extraction target content from which the important word originated, the information processing apparatus 1 selects the most important word having high relevance to the content of the target content. Can be extracted.
Further, since the important word and the most important word extracted by the above processing are language expressions divided in units of words called words, it is very easy for the user to understand and can be used easily and effectively as a search key. Therefore, content can be managed very efficiently by including in the metadata information that is easy for the user to recognize and easy to search.

Further, the information processing apparatus 1 selects the important word extraction target content from the content created or updated within a predetermined time before and after the creation or last update date and time of the target content. In content created or updated at a time away from the creation or last update date of the target content, keywords and similar words may be used with different meanings and nuances.
The information processing device 1 eliminates content that is temporally separated from the target content from the processing target, thereby substantially eliminating the content that deviates from the content of the target content, and is importantly related to the field of the target content. Since the word extraction target content is selected, more appropriate words can be extracted.

In addition, the information processing apparatus 1 can select important word extraction target content from pre-registered content, and can also select important word extraction target content from content included in the browser history. You can also choose. By selecting the important word extraction target content from the web content browsed by the user using the information processing apparatus 1, the content that has attracted the user's interest in the past is used as the important word extraction target content. By reflecting the user's taste in the selection of the important word extraction target content,
An important word extraction target content that is surely related to the content of the target content can be selected.

In the above embodiment, the configuration shown in FIG. 1 is realized mainly by a hardware configuration or by cooperation of hardware and software, and a specific form of each part constituting the information processing apparatus 1 The physical aspect is arbitrary.
In the above embodiment, an example in which a word is used as a keyword and an important word and a most important word are extracted and processed has been described, but the present invention is not limited to this, and a phrase including a plurality of words, Alternatively, clauses, sentences, paragraphs, etc. may be used as keywords, and information consisting of the language units can be extracted and processed instead of important words and most important words.

In the above embodiment, the case where the user operates the input unit 6 to input the field of the target content has been described. For example, the target content storage unit 55 stores the target content before executing the metadata generation process described above. If the metadata of the target content already contains information indicating the field of the target content, the input by the input unit 6 may be omitted by reading this information. Specifically, for example, information indicating the genre of the program is added to the digitally broadcast content, but this information may be used as information indicating the field of the target content.

Furthermore, although the case where the field of the target content is used as an example of the attribute information regarding the content of the target content has been described in the above embodiment, the present invention is not limited to this, and the attribute information regarding the content of the target content As a still image taken by myself,
Information indicating the origin of content such as purchased still images, theatrical movies, and TV broadcast videos, information indicating the country or region in which the target content was produced, and restrictions on broadcasting or broadcasting the target content such as R designation When the target content is a video, information indicating a person or an object, a location, or a landscape shown in the video can be used.

Specifically, the information processing apparatus 1 described in the above embodiment is a personal computer, P
The present invention can be applied to DAs, mobile phones, etc., but can be mounted or connected to digital still cameras, digital video cameras, scanners, projectors, televisions, printers, and other electronic devices that process image data. In this case, for example, if the present invention is applied to an image generation device such as a digital still camera, a digital video camera, and a scanner, metadata including feature information is added to the image as the target content is generated. Can do. Furthermore, if the present invention is applied to an image output device such as a projector, a television, or a printer, the feature information can be included in the metadata when outputting the image as the target content, and the metadata is output together with the image. be able to.
In addition, the specific detailed configuration of each unit constituting the information processing apparatus 1 can be arbitrarily changed without departing from the gist of the present invention.

It is a block diagram which shows the structure of the information processing apparatus which concerns on embodiment of this invention. It is a figure which shows the example of the keyword according to field. It is a figure which shows the example of registration content information typically. It is a flowchart which shows a metadata production | generation process. It is a flowchart which shows an important word extraction process. It is a flowchart which shows the most important word extraction process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Information processing apparatus, 2 ... Control part (Reference content acquisition means, extraction means, additional information generation means, reference expression acquisition means, candidate extraction means, content evaluation means, feature information extraction means, external content display means), 3 ... ROM, 4 ... RAM, 5 ... storage unit, 6 ... input unit, 7 ... output unit,
DESCRIPTION OF SYMBOLS 8 ... External interface, 9 ... Communication interface, 15 ... Communication line, 20 ... Network, 51 ... Application program, 52 ... Metadata generation program, 53 ... Data storage part for applications, 54 ... Keyword information storage part by field (standard expression) Storage means)
55 ... target content storage unit, 56 ... registered content information storage unit, 57 ... important word storage unit, 58 ... most important word storage unit.

Claims (11)

Reference content acquisition means for acquiring attribute information related to the content of the processing target content and acquiring reference content related to the attribute information;
Extraction means for extracting feature information from the reference content acquired by the reference content acquisition means;
Additional information generating means for generating additional information of the processing target content including the feature information extracted by the extracting means;
An information processing apparatus comprising: Reference expression storage means for storing a reference expression for each attribute information;
Reference expression acquisition means for acquiring a reference expression corresponding to the attribute information of the processing target content from the reference expression storage means,
The reference content acquisition means acquires the reference content including a reference expression acquired by the reference expression acquisition means or an expression similar to the acquired reference expression;
The information processing apparatus according to claim 1. The extraction means includes
Candidate extraction means for extracting candidate feature information from the plurality of reference contents when a plurality of the reference contents are acquired by the reference content acquisition means;
Content evaluation means for weighting each of the reference contents based on the appearance state of the reference expression;
Feature information extracting means configured to extract the feature information from a plurality of feature information candidates based on the weight of the reference content from which the feature information candidates have been extracted;
The information processing apparatus according to claim 2. The candidate extraction unit divides the language expression included in the reference content into language units, weights each language unit, and selects a part of the language expressions based on the weight in the language unit. A candidate for the feature information,
The information processing apparatus according to claim 3. The reference expression is a linguistic expression composed of a linguistic unit of a word, a phrase, a clause, a sentence, or a paragraph;
The information processing apparatus according to claim 2, wherein: An input means for inputting the attribute information;
The reference content acquisition means acquires the reference content related to the attribute information input by the input means;
The information processing apparatus according to claim 1, wherein: The reference content acquisition means acquires the reference content created or updated within a predetermined time from the date and time when the processing target content was created or updated;
The information processing apparatus according to claim 1, wherein: Comprising a communication means for transmitting and receiving information via an externally connected communication line;
The reference content acquisition means acquires the reference content from an external device communicable by the communication means;
The information processing apparatus according to claim 1, wherein: External content display means for displaying the content received by the communication means,
The reference content acquisition means acquires the content previously displayed by the external content display means as the reference content;
The information processing apparatus according to claim 8. Acquire attribute information regarding the content of the processing target content, acquire reference content related to the attribute information, extract feature information from the acquired reference content, and add additional information of the processing target content including the extracted feature information Generating,
An information processing method characterized by the above. Computer
Reference content acquisition means for acquiring attribute information related to the content of the processing target content and acquiring reference content related to the attribute information;
Extraction means for extracting feature information from the reference content acquired by the reference content acquisition means;
Additional information generating means for generating additional information of the processing target content including the feature information extracted by the extracting means;
Program to make it function.

Images