JP2005202485A - Video presenting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a video presenting device for improving the matching precision of program keywords and user taste keywords. <P>SOLUTION: This video presenting device is provided with a meta data storing part 101 which inputs a video to which attribute information is attached, a preference storage part 102 which stores user taste information, a preference extending part 104 which extends the user taste information by synonyms, a matching deciding part 106 which performs matching between the attribute information and the extended user taste information, a recommended program deciding part 107 which recommends a program based on the matching result and a display part 108 which presents the recommended program to the user. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a video presentation apparatus that presents video in accordance with matching between metadata added to the video and user preference data.

  Conventionally, a method for matching metadata added to video and user preference data has been studied. For example, there is a technique for recommending a program by extracting a keyword from an EPG (Electric Program Guide) and matching it with user preference data (see Patent Document 1). In this method, keywords weighted according to user operations such as appearance frequency, viewing time, and recording reservation are stored as user preference data. However, the matching method used here is a complete match of keyword strings, and even for keywords with the same meaning, the difference between Kanji and Hiragana notation and synonyms are handled as separate keywords. there were.

  Also, user preference data composed of theme dictionaries and themes that combine multiple keywords are defined in advance, and recommended programs are calculated by the product of thematic score and user preference data for keywords extracted from EPG program descriptions. There is a technique for selecting (see Patent Document 2). According to this technique, different keywords can be matched as belonging to the same theme. However, in order to support kanji, hiragana notation, and aliases, a theme dictionary with a huge amount of data in which all of these are registered is required. Creating such a dedicated dictionary requires high costs. Even if such a dictionary can be created, it cannot correspond to synonyms that have similar meanings but are not identical. In addition, since the matching process is compared with all registered keywords, there is a problem that it takes a long processing time.

In addition, there is a technique in which a plurality of user profiles such as genres, program titles, performers, etc. are prepared, and the user can select which profile to use when recommending a program. (See Patent Document 3). According to this technology, it is possible to recommend a program that more closely reflects the user's preference, but it is not possible to deal with anything other than a profile prepared in advance, and the operation of selecting a profile is complicated for the user. there were.
Japanese Patent No. 3351058 JP 2002-320159 A JP 2002-142160 A

  In the conventional matching method between the metadata added to the video and the user preference data, matching with synonyms could not be performed. Further, in order to perform matching in accordance with the user's situation, it is complicated because the user must explicitly indicate the items that are important.

  Therefore, the present invention provides a video presentation device that improves matching accuracy by expanding the keywords constituting metadata and user preference data with synonyms in matching metadata added to video and user preference data. For the purpose.

  The present invention includes means for inputting a video to which attribute information is added, user preference storage means for storing user preference information, information expansion means for expanding the user preference information with synonyms, the attribute information, and extended user preference. Provided is a video presentation device comprising means for matching information, video specifying means for specifying a video based on the matching result, and means for presenting the specified video to a user To do.

  If the video presentation device using the user preference matching by the synonym of the present invention is used, it is possible to detect a synonym even if the keyword representing the user preference and the keyword of the program do not exactly match. Can recommend well. Furthermore, it is possible to cope with dynamically changing user preferences by sensing user preferences with a sensor and dynamically changing a keyword representing the user preferences.

  Embodiments of the present invention will be described below with reference to the drawings.

(First embodiment)
FIG. 1 is a schematic configuration diagram of a video presentation device according to the first embodiment of the present invention. According to this apparatus, a program guide receiving unit 100 that receives a program guide is connected to a metadata storage unit 101 that extracts and stores keywords (attribute information) from the program guide. The preference storage unit 102 stores user preferences (preferences) as keywords, and is connected to the preference extension unit 104 together with the synonym dictionary 103 storing the synonyms. The preference extension unit 104 has a function of extending a preference by adding a synonym to the preference. In the preference extension unit 104, the output unit of the preference extension unit 104 is connected to an extended preference storage unit 105 that stores the extended preference. The output unit of the extended preference storage unit 105 is connected to the matching determination unit 106 together with the output unit of the metadata storage unit 101. The matching determination unit 106 compares the metadata and the extended preference to calculate the score of each program. The output unit of the matching determination unit 106 is connected to a recommended program determination unit 107 that determines a program recommended to the user according to the determination result. The recommended program determining unit 107 determines a recommended program according to the output data of the program guide receiving unit 100 and the matching determining unit 106, and outputs the program guide of the program to the display unit 108 for presentation to the user.

  The program receiving unit 100 is composed of, for example, an EPG compatible TV tuner, and extracts a program guide described in EPG from the received radio waves. FIG. 2 shows an example of a part of the program table, and one piece of program information includes items of an identifier, a broadcasting station name, a start time, an end time, a program title, and program contents. The title and program content are described by words or sentences.

  The metadata storage unit 101 is composed of, for example, a hard disk, extracts words from the received program guide, and stores the program guide. Common methods such as keyword matching and syntax analysis are used to extract words. FIG. 3 shows a state where the metadata storage unit 101 stores program information, and the program information includes an identifier, a program title, and a keyword extracted from the program content.

  The preference storage unit 102 is composed of, for example, a hard disk, and stores keywords that the user is interested in. FIG. 4 shows an example of a user preference. Two types of keywords, a keyword related to a preferable program and a keyword (preference) related to an unfavorable program, are stored in the preference storage unit 102. “Shopping” and “cooking” are registered as preferable keywords, and “baseball” is registered as an unfavorable keyword (non-preference). It is assumed that the keyword stored at the top is more interested in the user. In this example, “shopping” is more interested than “cooking”. It is assumed that the user can register and delete any keyword of interest.

  The synonym dictionary 103 is a general-purpose ontology dictionary composed of, for example, a hard disk and connected to a tree structure according to meaning, for example, an EDR electronic dictionary and a classification lexicon of Japan Electronic Dictionary Research Institute. You can use [National Institute of Japanese Language 1964] and Japanese vocabulary system [Ikehara et al. 1997].

The preference extension unit 104 is configured by, for example, an electronic circuit, and adds a synonym searched using the synonym dictionary 103 to the keyword stored in the preference storage unit 102. Here, only synonyms with different expressions but the same meaning are added. In the search for synonyms, for example, the degree of similarity having a value between 0 and 1 is calculated for all the words registered in the dictionary, and a keyword having this degree of similarity of 1 is added. Similarity calculation is performed, for example, by Leacock et al. (Leacock, C. and Chodorow, M. (1998). “Combining Local Context and WordNet Similarity for Word Sense Idenfication.” In Fellbaum, C. (Ed.), WordNet, An Electronic Lexical Database, pp.265-283. MIT Press). Here, w1 and w2 are keywords for calculating similarity, N is the number of nodes in the path p of the tree structure constituting the ontology dictionary from w1 to w2, and D is the deepest length of the dictionary tree.

  FIG. 5 is a result of searching for synonyms of keywords stored in the preference storage unit 102, “shopping” is searched as a synonym, and “cooking” is searched for “cooking” and “cooking”. ing. The preference extension unit 104 adds the searched synonym to the user preference. FIG. 6 shows an example of user preferences expanded with synonyms, and keywords are added in order from the first keyword stored in the preference storage unit 102. Taking “shopping” as an example, the synonym is “shopping”, and this keyword is added after “shopping”.

  The extension preference storage unit 105 is configured by, for example, a hard disk, and stores the result of the preference extension unit 104 extending the keywords stored in the preference storage unit 102. In this example, the extension preference shown in FIG. 6 is stored. The keyword filled in gray in the figure indicates an expanded preference.

  The matching determination unit 105 includes, for example, an electronic circuit, and compares a keyword representing program information stored in the metadata storage unit 101 with a keyword representing user preferences stored in the extended preference storage unit 105. Scores are calculated and priorities of programs to be recommended to the user are given.

  FIG. 7 shows the keyword scores used to calculate the score of the program, and the keywords filled in gray in the figure indicate expanded preferences. For the preferred keyword, the end of the preferred keyword stored in the expanded preference storage unit 105 is 1, and 1 is added to each non-synonymous word that moves in the beginning direction to increase the score of the first keyword. Similarly, with respect to unfavorable keywords, the unfavorable keyword stored in the expanded preference storage unit 105 has the end of −1 and 1 is subtracted for each non-synonym that moves in the head direction, so that the score of the leading keyword is the lowest. To do.

  FIG. 8 shows an example of calculating the score of a program. When the keyword representing the program information stored in the metadata storage unit 101 matches the keyword representing the user preference stored in the extended preference storage unit 105. A score shown in FIG. 6 is attached, and 0 is added when they do not match. For example, in the case of the program information with the identifier 4, since the keyword “tonight” is not in the table shown in FIG. 7, the score is 0, and similarly, the “baked fish” is also 0. “Cooking” exists in the table shown in FIG. The program information score is obtained by summing up the keyword scores. In this example, the scores of identifiers 3, 4, 5, and 6 are -1, +1, 0, and +2, respectively.

  As described above, the keyword “shopping” stored in the preference storage unit 102 and the “shopping” stored in the metadata storage unit 101 do not match as they are, but are matched by expanding the preference with synonyms. Will be able to.

  The recommended program determination unit 107 is constituted by, for example, an electronic circuit, compares the program scores calculated by the matching determination unit 105, and selects the program with the highest score. In the example of FIG. 8, since the program with the highest total score is the program with the identifier 6, the recommended program determination unit 107 selects the identifier 6 and displays the program information with the identifier 6 input from the program guide receiver 100. To 108.

  The display unit 108 is configured by a display, for example, and displays the program information selected by the recommended program determination unit 107 on the screen and recommends it to the user. FIG. 9 shows an example of a screen for recommending the program with the identifier 6.

  FIG. 10 is a flowchart for explaining the above processing procedure. According to this, first, the program receiving unit 100 receives program information (step S900). The metadata storage unit 101 extracts keywords from the program guide by syntax analysis and stores them (step S901). The preference expansion unit 104 expands the synonym dictionary 103 by adding synonyms to the keywords stored in the preference storage unit 102 (step 902).

  The matching determination unit 106 calculates the score of the program stored in the metadata storage unit 101 (step 903). The recommended program determination unit 107 selects the program with the highest score (step S904), and displays the program information on the display unit 108 (step S905).

  FIG. 11 is a flowchart for explaining a processing procedure for calculating a program score in step 903. This flowchart represents a procedure for calculating the score of one program. According to this, first, the top keyword associated with the program stored in the metadata storage unit 101 is selected (step S1000). It is determined whether there is a keyword that matches the selected keyword among the keywords stored in the extended preference storage unit 105 (S1001). If this determination is YES, processing proceeds to step S1002. If the determination is NO, the process proceeds to step S1003.

  If a matching keyword is found, a score corresponding to the keyword is added (step S1002). Next, it is determined whether all keywords have been checked (S1003). If this determination is YES, the process proceeds to step 1004, and if not, the process returns to step S1000 to check the next keyword. In step S1004, the total score is output. This process is repeated for each program to calculate the scores of all programs.

  If the first embodiment described above is used, a synonym can be detected even when the same keyword as the keyword of the user preference is not present in the program information, so that the program can be recommended with high accuracy.

  In the first embodiment, the program receiving unit 100 receives the program guide by radio waves. However, the present invention is not limited to this, and any means can be used as long as it can receive the program guide. For example, iEPG (Internet EPG) may be received over the Internet. In addition, the preference extension unit 104 calculates the similarity by the method of Leacock et al. However, the present invention is not limited to this, and any method can be used as long as it can calculate the similarity between keywords. Furthermore, although only the synonym is added in the preference extension unit 104, the present invention is not limited to this, and a threshold value may be set in advance, and a keyword having a similarity degree exceeding the threshold value may be added. In addition, although only one recommended program determining unit 107 is selected, the present invention is not limited to this, and the order may be given and the user may browse the display unit 108 in order.

  Furthermore, in the said embodiment, although user preference information is expanded with a synonym, you may extend attribute information with a synonym. In this case, matching is performed between the extended attribute information and the user preference information. FIG. 28 shows keyword scores when user preferences are not expanded, and the method of assigning keyword scores is the same as in FIG. FIG. 29 shows the program information expanded with synonyms, and the synonym of the preference is inserted after one preference in the same way as the user preference is expanded in FIG. The gray preference in the figure is an expanded preference. FIG. 30 shows how the program score is calculated using the program information score table shown in FIG. 28 and the extended user preferences shown in FIG. 29. Gray program information is expanded. It shows that. As a result, “baseball” is −1, “cooking” is +1, and “shopping” is +2. As a result, the total scores of the programs with identifiers 3, 4, 5, and 6 are −1, +1, 0, and +2, respectively. Calculated. Since the total score of the program with the identifier 6 is the largest, the recommended program determination unit 107 selects the identifier 6.

  In the above embodiment, the matching determination unit 105 performs the matching determination based on whether or not the keywords completely match. However, the similarity between the keyword and the user preference may be used. FIG. 31 is a diagram for explaining a program score calculation method using similarity, and uses the same attribute information as the program score calculation method based on complete keyword matching shown in FIG. For example, the keyword “grilled fish” has a similarity of 0.6 to the user preference “dish” with a word score of 3, and is represented as “3 × 0.6” in the figure. In this example, the score is calculated by calculating the word similarity between each keyword and each user preference, and the highest score among them is used for program score calculation. Even if the similarity with the user preference “Cooking” is 0.4, the word score is “2 × 0.4 = 0.8”, and the similarity with the user preference “Cooking” is “ Not adopted because it is smaller than 1.8 ". The program score is calculated as the sum of the word scores thus calculated. In this example, the maximum value of the word similarity is adopted, but not limited to this, the sum of the similarities may be used. In this example, the degree of similarity is obtained between the attribute information and the extended user information. However, the present invention is not limited to this, but between the extended attribute information and the user information, or between the attribute information that is not expanded and the user information that is not expanded. You may use the similarity calculated between.

(Second Embodiment)
In the first embodiment, matching keywords are improved by expanding keywords representing user preferences with synonyms. However, in the second embodiment, a user's situation is captured by a sensor to correspond to the user situation. Add keywords to expand user preferences. In this embodiment, the preference expansion unit 104 in the first embodiment expands a keyword according to the user situation instead of expanding the keyword stored in the preference storage unit 102 with a synonym.

  FIG. 12 is a schematic configuration diagram of a video presentation device according to the second embodiment. This apparatus is configured by adding a status information acquisition unit 109 and a status preference storage unit 110 to the schematic configuration of the first embodiment.

  The user status information acquisition unit 109 includes, for example, a person identification sensor and a blood glucose level sensor attached to the user, and acquires the person watching the television and the blood glucose level of the user. The person identification sensor is composed of, for example, a wireless ID tag incorporated in a badge and a wireless ID reader incorporated in a television, and identifies a person around the television. The blood glucose level sensor is composed of, for example, a portable optical blood glucose level measuring device and a wireless device, and reads the blood glucose level of the user wearing these. The data read by the sensor is classified into meaningful units and assigned a sensor ID.

  FIG. 13 is an example of association between sensor data and sensor ID, and sensor ID 1 represents that the user is eating. For example, any one of the users exceeds the threshold value of the blood glucose level sensor. If so, determine that you are eating. The sensor ID 2 represents that the user A is watching the television. For example, it is determined that the ID read by the wireless ID tag includes the ID of the user A. Similarly, sensor ID3 represents that user B is watching TV.

  The user situation information acquisition unit 109 outputs True if the sensor ID satisfies the sensor condition, and outputs False if not. In the illustrated example, only sensor ID2 satisfies the sensor condition, and other IDs do not satisfy the sensor condition. The situation preference storage unit 110 is composed of, for example, a hard disk, and stores keywords related to the state of the sensor ID.

  FIG. 14 is an example of correspondence between sensor IDs and keywords. The sensor ID 1 indicates that a meal is in progress, and the keywords “vomiting” and “blood” are registered as undesirable in the meal. The sensor ID 2 represents the preference of the user A, and “music” is registered as a preferable keyword, and “cooking” is registered as an undesirable keyword. Similarly, the sensor ID 3 represents the preference of the user B, and “baseball” is registered as a preferable keyword.

  The preference expansion unit 104 according to the second embodiment refers to the state of the sensor ID, adds and deletes a keyword having a sensor ID of True to the storage information in the preference storage unit 102, and sets user preference information. Update. In this example, among the sensor IDs shown in FIG. 13, the state of the sensor ID2 is True, indicating that the user A is watching the television. In this case, synonyms are searched for the keyword associated with the sensor ID 2 as shown in FIG. The synonym search is performed in the same manner as in the first embodiment, and “music” and “easy” are searched as synonyms for “music”, and “cooking” and “cooking” are searched as synonyms for “cooking”. Is done.

  Next, in order to prevent a keyword to be added as a preferred keyword or its synonym from being registered as an unfavorable keyword or a keyword to be added as an unfavorable keyword or its synonym from being already registered as a preferred keyword, FIG. The keyword and its synonym associated with the searched sensor ID 2 are deleted from the user preferences. In this example, “cooking” of the user preferences shown in FIG. 4 matches the keyword searched in FIG. Thereafter, the keyword associated with the sensor ID 2 is added to the head of the preference.

  FIG. 16 shows user preferences expanded in this manner, with “food” being deleted and “cooking” being added as an unfavorable keyword at the top. “Music” is simply added to the top of the preferred keyword because there was no keyword to delete.

  If there are a plurality of sensors whose status is True, the same processing is repeated from the sensor ID with the smallest sensor ID, and finally one expanded user preference is generated. Since the subsequent processing is the same as in the first embodiment, the description thereof is omitted, but the music program with ID4 shown in FIG. 2 is finally recommended to the user.

  FIG. 17 shows a flowchart of processing for extending user preferences in the preference extension unit 104 of the second embodiment. According to this, first, 0 is set to the variable n for storing the sensor ID (step S1700), and 1 is added to the variable n so that all sensor IDs can be repeatedly processed (step S1701).

  Next, it is checked whether or not the state of the sensor ID n is True (step S1702). If True, the process proceeds to step S1703, and if not, the process proceeds to step S1706. In step S1703, the synonym of the keyword registered in association with the sensor with the sensor ID n is searched (step S1703), and the registered keyword and the searched synonym are deleted from the user preference (step S1705). Thereafter, the registered keyword is added to the head of the user preference to generate an extension preference (step S1705). In order to perform this process for all sensors, it is compared whether or not the variable n is smaller than the number N of sensors. If the variable n is small, the process returns to step S1701 to perform the process for the next sensor. The processing is stopped (step S1706).

  According to the second embodiment described above, the user preference can be dynamically changed according to the user's situation, so that a program according to the user's situation can be recommended.

  In the second embodiment, the preference extension unit 104 deletes not only the keyword stored in the environment preference storage unit 110 but also its synonym from the keyword stored in the preference storage unit 102. Not only the keywords stored in the environment preference storage unit 110 may be processed. Further, the processing after the matching determination is the same as that in the first embodiment. However, the present invention is not limited to this. For example, in the second embodiment, the user preference expanded by the environment information shown in FIG. 16 is further expanded with synonyms as in the first embodiment, but the matching determination is performed without expanding with the synonyms. Also good.

(Third embodiment)
In the first embodiment, keywords representing user preferences are expanded by synonyms and program information matching is performed. In the third embodiment, partial videos are similarly hidden or highlighted by expansion by synonyms of user preferences. Display control. In this display control, for example, a partial video unpleasant to the user is automatically hidden or a partial video that is strongly interested is emphasized.

  In the description of the embodiment, for example, a moving image displayed on a television or the like is referred to as a “video”, and a meaningful part such as a person or an object in the video is described as a separate unit from the whole video. It will be called a “video object”. The video object may be a partial image, or may be metadata in which a portion in the video is represented by, for example, coordinate values. Here, the latter metadata will be described as being used.

  In the present embodiment, the program guide receiving unit 100 in the first embodiment receives a video and a video object instead of receiving a program guide, and the recommended program determining unit 107 displays a video object instead of determining a recommended program. There is only a difference between determining the method and displaying the combined video information by hiding or highlighting the video object instead of displaying the program information recommended by the display unit 108 to the user. explain.

  FIG. 18 is a schematic configuration diagram of a video presentation device according to the third embodiment. This apparatus uses a video receiving unit 111 instead of the program receiving unit 100, a video object display method determining unit 112 instead of the recommended program determining unit 107, and a display unit in the schematic configuration of the first embodiment. The video display unit 113 is used instead of the video display unit 113.

  The video receiving unit 111 is configured by, for example, an EPG (Electric Program Guide) compatible TV tuner, and receives a video signal and a video object signal. FIG. 19 shows an example of a received signal. In this example, a video showing a butterfly, a snake, and a cockroach is received as the video 1900, and video objects are received as ellipse graphic information 1901, snake graphic information 1902, and cockroach graphic information 1903 surrounding the partial video of the butterfly. To do. In this figure, the graphic information is indicated by broken lines for the sake of explanation, but this broken line is not displayed on the actual screen. The video signal and the video object signal are transmitted from the broadcast station once every 1/30 seconds, and the video signal is input as video data for one frame in one reception. Shall be received.

  FIG. 20 shows an example of a video object signal (metadata) input in one reception, and the video object signal (metadata) is stored in the metadata storage unit 101 every time a video signal is received. In this example, the metadata includes a video object identifier 2000, video object graphic information 2001, and a keyword 2002 describing the video object. The graphic information 2001 is composed of the center coordinates of the ellipse, the minor axis, the major axis, and the inclination. The keyword 2002 is composed of a sequence of words. In this example, the video object with the identifier 134 is “snake”, the video object with the identifier 135 is “cockroach” and “insect”, and the video object with the identifier 136 is “butterfly”. , “Insect” is added as a description of the video object.

  FIG. 21 shows user preference information stored in the user preference storage unit 102. In this example, it is assumed that “insect” is stored as a preferred keyword, and “oilworm” and “snake” are stored as undesirable keywords.

  FIG. 22 shows an example of the contents stored in the extended preference storage unit 105. When the apparatus receives the video signal, the preference extension unit 104 extends the user preference information with synonyms as in the first embodiment. In this example, “oilworm” is added with another name “cockroach” and another notation “aphid”.

  FIG. 23 shows an example of a user preference score table created by the same method as that of the first embodiment for the matching determination by the matching determination unit 106.

  FIG. 24 shows an example of the result of calculating the score of each video object by the same method as in the first embodiment in order to make a matching determination by the matching determination unit 106. The video object display method determination unit 112 receives the score calculated by the matching determination unit 106 and determines whether to display the video object as it is or to paint it. Here, it is determined that a video object having a negative score is not performed, and a video object having a negative score is determined to be filled.

  FIG. 25 shows an example in which a video and a video object are combined and displayed according to the determination of the video object display method determination unit 112. In this example, since the score of the video object “snake” with the identifier 134 and the video object “cockroach” with the identifier 135 are negative, they are displayed in a solid color. The fill is performed in an elliptical shape using graphic information representing the video object received by the video receiver 111. Since the video object of “Chow” with the identifier 136 is positive, it is displayed as it is.

  FIG. 26 shows a flowchart of the above-described processing procedure. According to this, first, the video receiver 111 receives a video and a video object (step S2600). The metadata storage unit 101 stores the video object as graphic information and a keyword (step S2601). The preference extension unit 104 extends the keywords stored in the preference storage unit 102 with synonyms (step S2602). The matching determination unit 106 calculates the score of the video object stored in the metadata storage unit (step S2603). The video object display method determination unit 112 refers to the score of the video object and determines to display the video object having a negative score (step 2604). The display unit 113 fills the video object determined to be filled with the graphic received by the video receiving unit 111, and displays the video object that is not so as it is (step S2605).

  The expansion of user-preferred keywords by synonyms takes time, but the matching method of this embodiment is suitable for high-speed matching on a frame basis because it can be prepared before viewing a program.

  If the third embodiment described above is used, even if the same keyword as the keyword of the user preference is not found in the keyword of the video object, it can be detected if it is a synonym. Can control.

  According to the third embodiment, the user preference keyword is set in advance by the user. However, the present invention is not limited to this, and the user preference keyword may be set dynamically with reference to the user's current preference. For example, similarly to the second embodiment, the user preference keyword may be dynamically changed by associating the sensor with the keyword. In addition, the video receiving unit 111 receives the video object information for each video frame, but the video receiving unit 111 is not limited thereto, and may receive it by thinning or prefetching. In the case of thinning, prediction may be performed from data received twice before, and in the case of prefetching, linear interpolation may be performed from previous and subsequent data.

  18 includes the function or situation information acquisition unit 109 and the situation preference storage unit 110 of the embodiment of FIG. 1 including the program receiving unit 100 and the recommended program determining unit 107 as shown in FIG. The functions of the embodiment or both functions may be added. According to this, in addition to performing display control for concealing or highlighting partial video by extension by user preference synonyms, even if the program keyword does not have the same keyword as the user preference keyword, , Can recommend the program with high accuracy. In addition, the user preference can be dynamically changed according to the user's situation, in addition to performing display control for concealing or highlighting the partial video by extension by a user preference synonym. It is possible to recommend programs according to the situation.

  In the above embodiment, user preference information is expanded with synonyms, but attribute information may be expanded with synonyms as in the first embodiment.

  In addition, this invention is not limited to the said embodiment, In the implementation stage, it can change variously in the range which does not deviate from the summary. Further, the above embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent elements are deleted from all the constituent elements shown in the embodiment, at least one of the problems described in the column of the problem to be solved by the invention can be solved, and is described in the column of the effect of the invention. When at least one of the effects is obtained, a configuration in which this configuration requirement is deleted can be extracted as an invention.

The block diagram of the image | video presentation apparatus according to 1st Embodiment of this invention. Figure showing an example of the program guide The figure which shows an example of the program information memorize | stored in a metadata memory | storage part Diagram showing an example of user preferences The figure which shows an example of a keyword synonym search result Diagram showing an example of user preferences extended with synonyms Figure showing keywords and scores The figure which shows an example which calculates a program score Figure showing the program recommendation screen The flowchart figure which shows the process of the video presentation apparatus of FIG. The flowchart figure which shows the process sequence which calculates the score of a program The block diagram of the image | video presentation apparatus according to 2nd Embodiment of this invention. The figure which shows an example of matching with sensor data and sensor ID The figure which shows an example of matching with sensor ID and a keyword Diagram for explaining synonym search Diagram showing expanded preferences The flowchart figure which shows the process of the video presentation apparatus of FIG. The block diagram of the image | video presentation apparatus according to 3rd Embodiment of this invention. Diagram showing video object Diagram explaining video object signal The figure which shows user preference information Diagram showing an example of user preferences extended with synonyms Figure showing an example of the score table The figure which shows an example of the result of calculating the score of a video object The figure which shows the state which displays the picture object selectively The flowchart figure which shows the process of the video presentation apparatus of FIG. The block diagram of the image | video presentation apparatus which included embodiment of FIG. 1 and / or embodiment of FIG. 18 in embodiment of FIG. Diagram showing user information and score not expanded Diagram showing extended program attribute information The figure which shows an example which calculates a program score using the extended attribute information and the user information which is not extended The figure which shows an example which calculates a program score using similarity

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Program schedule receiving part 101 ... Metadata storage part 102 ... Preference storage part 103 ... Synonym dictionary 104 ... Preference extension part 105 ... Extended preference storage part 106 ... Matching determination part 107 ... Recommendation Program determining unit 108 ... Display unit 109 ... Situation information acquisition unit 110 ... Situation preference storage unit 111 ... Video receiving unit 112 ... Video object display method determining unit 113 113 Video display unit

Claims (11)

Means for inputting video with attribute information;
User preference storage means for storing user preference information;
Information expansion means for expanding the user preference information with synonyms;
Means for matching between the attribute information and the extended user preference information;
Video specifying means for specifying a video based on the matching result;
Means for presenting the identified video to the user;
A video presentation apparatus comprising: Means for inputting video with attribute information;
User preference storage means for storing user preference information;
Information expansion means for expanding the attribute information with synonyms;
Means for matching between the extended attribute information and the user preference information;
Video specifying means for specifying a video based on the matching result;
Means for presenting the identified video to the user;
A video presentation apparatus comprising:   3. The information expansion means for expanding the user preference information with a synonym, wherein the matching means performs matching between the extended attribute information and the extended user preference information. Video presentation device. Status acquisition means for acquiring user status;
Situation keyword storage means for storing a keyword corresponding to the acquired situation;
In order to update the user preference information according to the situation acquired by the situation acquisition means, the apparatus further comprises situation keyword control means for adding or deleting the keyword to or from the user preference storage means. The video presentation apparatus according to claim 1 or 2.   The video presentation apparatus according to claim 1, wherein expansion by synonyms is limited to only synonyms.   The video presentation apparatus according to claim 1, wherein the matching unit uses a similarity between attribute information and user preference information. Means for inputting video and video objects;
User preference storage means for storing user preference information;
Information expansion means for expanding the user preference information with synonyms;
Means for matching between the video object and the extended user preference information;
Video processing means for processing the video object based on the matching result and generating a display video by combining the video object;
Means for presenting the display video to a user;
A video presentation apparatus comprising: Means for inputting video and video objects with attribute information added thereto;
User preference storage means for storing user preference information;
Information expansion means for expanding the attribute information with synonyms;
Means for matching between the extended attribute information and the user preference information;
Video processing means for processing the video object based on the matching result and generating a display video by combining the video object;
Means for presenting the identified video to the user;
A video presentation apparatus comprising:   9. The information expanding means for expanding the user preference information with a synonym, wherein the matching means performs matching between the extended attribute information and the extended user preference information. Video presentation device.   The video presentation apparatus according to claim 75, further comprising video identification means for identifying a video object based on the matching result, wherein the presentation means presents the identified video object to a user.   Status acquisition means for acquiring user status, status keyword storage means for storing a keyword corresponding to the acquired status, and the keyword in the user preference storage means for updating the user preference information according to the acquired status 7. The video presentation apparatus according to claim 5, further comprising status keyword control means for adding to or deleting from the status keyword control means.

Images