JP7134662B2 - Information delivery system and information delivery method - Google Patents

Description

The present invention relates to an information distribution system, an information distribution method, and a program.

In recent years, methods for distributing information using networks have diversified. For example, there is known a technique of distributing advertisements according to the content of conversation by analyzing voice data during a videophone call (see, for example, Patent Document 1).

As another information distribution method, a method of distributing live moving images using a live distribution service is known. A live distribution service is a service for distributing moving images captured by a user to other users via the Internet at the same time as the images are captured. The live distribution service is capable of distributing various moving images shot by users, for example, distributing moving images shot for advertisement purposes. This live distribution service also allows viewing users to send comments. For this reason, two-way communication between users who post videos and users who view them is possible, and is attracting attention as a new communication method.

JP 2009-238199 A

However, live distribution services do not allow users to view live videos at any time, and in many cases the period during which users can view live videos is limited to when the live videos are being distributed. In this case, the timing of recommending a live video to the user is limited to the time when the live video is being distributed, and it is difficult to efficiently recommend the live video to the user who is unaware of the live video.

The present invention has been made in consideration of such circumstances, and provides an information distribution system, an information distribution method, and a program capable of efficiently recommending live moving images using a live distribution service. One of the purposes is to

One aspect of the present invention includes an analysis unit that analyzes information about a live video being distributed, a generation unit that generates reference information based on the analysis result of the analysis unit, and a request sent from a terminal device of a service user. an acquisition unit that acquires information, and referring to the reference information generated by the generation unit, based on the request information that is acquired by the acquisition unit, the live video being distributed is presented to the service user. and a distribution unit that distributes information about the live video determined by the determination unit to the service user.

According to one aspect of the present invention, a live video can be efficiently recommended using a live distribution service.

1 is a schematic diagram of an information distribution system 1 according to an embodiment; FIG. 1 is a configuration diagram of an information distribution system 1 according to an embodiment; FIG. 7 is a diagram showing an example of the contents of first reference information 731. FIG. 7 is a diagram showing an example of the contents of second reference information 732. FIG. 7 is a diagram showing an example of the contents of third reference information 733. FIG. 8 is a diagram showing an example of contents of user information 834. FIG. 8 is a diagram showing an example of contents of ranking information 835. FIG. 4 is a sequence diagram showing processing in the information distribution system 1 according to this embodiment; FIG. 4 is a flow chart showing an example of processing executed in the live distribution server 100. FIG. 7 is a flow chart showing an example of processing executed in the index server 700. FIG. 8 is a flow chart showing an example of processing executed in a recommendation distribution server 800. FIG. 8 is a flow chart showing an example of processing executed in a recommendation distribution server 800. FIG.

Hereinafter, embodiments of an information distribution system, an information distribution method, and a program according to the present invention will be described with reference to the drawings.

FIG. 1 is a schematic diagram of an information distribution system 1 according to an embodiment. The information distribution system 1 includes, for example, a live distribution server 100, one or more terminal devices 200, one or more terminal devices 300, an information providing server 400, one or more terminal devices 500, an analysis server 600, an index It includes a server 700 and a recommendation distribution server 800 . These configurations are connected to each other by a network NW and communicate with each other via this network NW. The network NW includes, for example, a part or all of a WAN (Wide Area Network), a LAN (Local Area Network), the Internet, a leased line, a wireless base station, a provider, and the like.

The terminal devices 200, 300, and 500 are portable terminal devices having at least a communication function and a display function, such as a mobile phone such as a smart phone, a tablet terminal, and a PDA (Personal Digital Assistant).

The terminal device 200 is used, for example, by a live contributor A who posts a live video to the live distribution server 100 . The terminal device 300 is used, for example, by a live viewer B who receives a live video distributed by the live distribution server 100 . The terminal device 500 is used, for example, by a service user C who uses an information providing service.

An information providing service is, for example, a service provided by the information providing server 400 . The terminal device 500 transmits request information to the information providing server 400, and displays information received from the information providing server 400 (hereinafter referred to as provided information) on its own display unit. Note that the terminal device 500 may display the provided information using a web browser or an application program.

The request information is, for example, information requesting transmission of information other than live video. The request information includes information about the service user C, information about the content of the provided information, and the like. Information about service user C includes attributes of service user C (age, sex, region, occupation, etc.), identification information (user ID), current location, current time, and the like. Note that the user ID may be registered in the information providing server 400 in association with other information about the service user C (for example, information about the hobbies and preferences of the service user C). Other information about service user C may also be included in the request information. The information about the contents of the provided information includes queries, keywords included in the provided information, information about the keywords, and the like.

The live distribution server 100 uses live moving images to provide, for example, a service that enables two-way communication. For example, the live distribution server 100 distributes in real time the live video posted by the live poster A using the terminal device 200 . Specifically, the live distribution server 100 distributes the video shot by the live poster A at a timing close to the shooting timing (a certain degree of lag due to communication delay and buffering), for example, It is delivered to the terminal device 200 and the live viewer B's terminal device 300 . The live viewer B can operate the terminal device 300 while viewing the live video distributed from the live distribution server 100 to post message information including comments and illustrations. Note that the message information is an example of posted information. The live distribution server 100 reflects the message information posted by the live viewer B on the live video, and distributes the live video to the live poster A and the live viewer B. Thereby, the live contributor A can obtain the reaction of the live viewer B in real time. In this manner, the live distribution server 100 can realize two-way communication between the live contributor A and the live viewer B. FIG.

The live videos posted by the live contributor A include, for example, videos related to advertisements for products and services. Live contributor A takes a picture of himself explaining things related to the target product (for example, features of the product, how to use it, impressions, etc.) using the terminal device 200, and posts the taken video. The target product may be a product handled by the live contributor A, or may be a product handled by another person. Note that live videos are not limited to videos related to advertisements, and may include various videos.

The information providing server 400 receives the request information from the terminal device 500 and transmits the provision information to the terminal device 500 that transmitted the request information. The provided information is, for example, information corresponding to the request information. The provided information includes, for example, web pages, information provided by application software, and other information that can be displayed on the terminal device 500 . Information included in these pieces of provided information is hereinafter referred to as content. For example, the information providing server 400 performs a search based on a query specified by the service user C and transmits the search result to the terminal device 500 . Further, the information providing server 400 may transmit the content of the address designated by the service user C to the terminal device 500, or may transmit information corresponding to the current location of the service user C to the terminal device 500. good. The information providing server 400 acquires request information for each service user C. FIG. The information providing server 400 may acquire information including a history of used queries, a history of browsed content, and the like as information based on the request information. The information providing server 400 may store the acquired request information (or information based on the request information) in its own storage unit, or may transmit it to the recommendation distribution server 800 .

The recommendation distribution server 800 determines a live video (hereinafter referred to as a push video) to be presented to the service user C from the live videos being distributed by the live distribution server 100 .

The information providing server 400 embeds a video with a link for jumping to the push video determined by the recommendation distribution server 800 as a recommendation in the content accessed by the service user C, and transmits the recommendation to the terminal device 500 . By doing so, the service user C can access the live distribution server 100 via the embedded recommendation and view the push video.

Next, the configuration of the information distribution system 1 according to the embodiment will be described with reference to FIG. FIG. 2 is a configuration diagram of the information distribution system 1 according to the embodiment.

(Live distribution server 100)
The live distribution server 100 includes, for example, a communication unit 110, a storage unit 130, and a processing unit 150. The communication unit 110 includes, for example, a communication interface such as NIC. The storage unit 130 is, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory such as an SSD, an HDD (Hard Disk Drive), or the like. The storage unit 130 includes, for example, a distribution folder 131, an archive folder 132, a message folder 133, and the like. Predetermined information is stored in each folder by the processing unit 150 . The storage unit 130 may be an external storage device such as a NAS (Network Attached Storage) that the live distribution server 100 can access via a network.

The distribution folder 131 is a folder for storing live videos posted by the live contributor A and to be distributed from now on. The archive folder 132 is a folder for storing already distributed live moving images. The message folder 133 is a folder for storing message information posted by the live viewer B and the like.

The processing unit 150 includes a setting unit 151 and a data management unit 152, for example. Some or all of these components are implemented by a processor such as a CPU (Central Processing Unit) executing a program (software) stored in the storage unit 130 . In addition, some or all of the functions of these components may be realized by hardware such as LSI, ASIC, FPGA (including circuitry), or by cooperation of software and hardware. may be implemented.

When receiving a distribution request from live poster A using communication unit 110, setting unit 151 sets a channel for distributing the live video posted by live poster A (hereinafter referred to as a distribution channel).

The data management unit 152 manages information about posted live videos. Information related to live moving images includes, for example, live data and message information. The live data includes, for example, moving image data and audio data included in the live moving image, meta information of the live moving image, and the like. For example, the data management unit 152 distributes posted live data and message information on a distribution channel. In parallel with this, the data management unit 152 transfers at least part of the posted live data and message information to the analysis server 600 .

Specifically, the data management unit 152 temporarily stores the live data received from the terminal device 200 in the distribution folder 131 . The data management unit 152 uses the communication unit 110 to read the video data and audio data of the live video stored in the distribution folder 131, and distributes the information read from the distribution folder 131 to the terminal device of the live poster A on the distribution channel. 200, to the terminal device 300 of the live viewer B of the distribution channel. In addition, when the service user C accesses the live distribution server 100 via the push video determined by the recommendation distribution server 800, the data management unit 152 also causes the terminal device 500 of the service user C to The information read from the distribution folder 131 is transmitted. By doing so, the live video is distributed on the distribution channel. In addition, the data management unit 152 transmits analysis target data (including, for example, audio data, message information, meta information, etc.) of the live data of the live moving image distributed on the distribution channel, excluding the moving image data, to the analysis server 600. do. The data management unit 152 moves the information on the live video distributed on the distribution channel to the archive folder 132 .

The data management unit 152 also stores the posted message information in the message folder 133 . The data management unit 152 reflects the received message information in the live data within the distribution folder 131 . For example, the data management unit 152 superimposes a comment or an illustration on the live video and displays it. After reflecting the message information in the live data, the data management unit 152 deletes the message information in the message folder 133 .

Further, when distributing a live video to the terminal device 500 of the service user C, the data management unit 152 arbitrarily determines the start time of the live video to be distributed, and distributes the live video from the determined start time from the archive folder 132. Information read out from the archive folder 132 may be transmitted. For example, the data management unit 152 determines N seconds before the conversion number spike, N seconds before the message number spike, N seconds before the viewer number spike, etc. as the start time. A spike includes, for example, a case where the rate of increase is equal to or higher than a threshold value, or a case where the value becomes several times (eg, 3 times or more) a value a predetermined time ago (eg, 5 minutes ago). Note that the start point is not limited to the timing when the spike occurs, and may be the timing when the number of conversions or the like begins to increase. By doing so, for example, since the live video can be reproduced from the timing that is likely to lead to conversion, effective advertisement distribution can be performed.

(Analysis server 600)
The analysis server 600 includes a communication unit 610, a storage unit 630, and an analysis unit 650, for example. The communication unit 610 includes, for example, a communication interface such as NIC. The storage unit 630 is, for example, a RAM, a ROM, a flash memory such as an SSD, an HDD, or the like. The storage unit 630 stores information required when the analysis unit 650 executes analysis processing, analysis results, and the like. The storage unit 630 may be an external storage device such as a NAS that can be accessed by the analysis server 600 via a network.

The analysis unit 650 analyzes the analysis target data received from the live distribution server 100 and transmits analysis results, meta information, and the like to the index server 700 .

The analysis unit 650 associates the analysis result with time information in the live video. The time information within the live moving image is information indicating the timing corresponding to the analysis result on the time axis within the live moving image. The time information in the live video may be, for example, information indicating the time at which the target video (video per unit time included in the live video) was distributed. may be information indicating the timing at which is reproduced using the elapsed time from the start of the live moving image.

For example, the analysis unit 650 converts the audio data included in the live moving image into text data by analyzing the audio data included in the live moving image. The analysis unit 650 transmits to the index server 700 information in which time information indicating the timing at which the text data appears is associated with the text data acquired by analyzing the voice data.

Also, the analysis unit 650 may analyze viewer attributes, viewer trends, and excitement trends based on meta information. The analysis unit 650 associates time information in the live video with the analysis result. The meta information includes, for example, the number of viewers of the live video, the number of conversions, attribute information about viewers of the live video, the number of recent views, the number of message information, and the like. The number of conversions includes, for example, the number of users who purchased products via live videos, and the number of users who jumped to product-related content (homepages, advertising pages, etc.) via live videos. . The most recent number of views is the number of viewers within the latest predetermined time during the delivery period of the live video.

For example, the analysis unit 650 may analyze viewer attributes based on live viewer B's attribute information. Viewer attributes include, for example, attribute tendencies of viewers viewing live moving images. Note that the analysis unit 650 may analyze viewer attributes in a past period within a predetermined time period from the current time based on time information in the live video, or instantaneous viewer attributes during distribution of the live video. good.

The analysis unit 650 may analyze viewer trends based on the number of viewers of the live video, the number of conversions, attribute information about viewers of the live video, the most recent number of views, and the like. Viewer trends include, for example, viewership ratings of live videos and their trends, purchase trends of viewers viewing live videos, and the like. The analysis unit 650 may analyze viewer trends for each viewer attribute.

The analysis unit 650 may analyze the excitement trend based on the number of recent views, the number of message information, and the like. The excitement trend includes, for example, the degree of excitement associated with time information in the live moving image. The degree of excitement increases, for example, as the number of viewers of the live video, the number of conversions, the number of recent views, the number of message information, and the like increase. The analysis unit 650 may analyze the excitement trend for each viewer attribute.

(Index server 700)
The index server 700 includes, for example, a communication unit 710, a storage unit 730, and a processing unit 750. The communication unit 710 includes, for example, a communication interface such as NIC. The storage unit 730 is, for example, a RAM, a ROM, a flash memory such as an SSD, an HDD, or the like. The storage unit 730 stores, for example, first reference information 731, second reference information 732, third reference information 733, and the like. The storage unit 730 may be an external storage device such as a NAS that the index server 700 can access via a network. The processing unit 750 generates first reference information 731 , second reference information 732 and third reference information 733 based on the analysis result by the analysis server 600 .

FIG. 3 is a diagram showing an example of the contents of the first reference information 731. As shown in FIG. The first reference information 731 is, for example, information in which a first related word is associated with time information. The first related word is a word included in the text data extracted by the analysis server 600 based on the voice data. The time information is, for example, information indicating the timing at which the first related word appears in the live moving image.

FIG. 4 is a diagram showing an example of the contents of the second reference information 732. As shown in FIG. The second reference information 732 is, for example, information in which set channels, meta information, poster attributes, viewer attributes, viewer trends, excitement trends, etc. are associated with live IDs. A live ID is identification information for identifying a live moving image. The set channel is information indicating the set channel assigned to the live video indicated by the live ID. The poster attribute is information indicating attributes of the live poster A, and includes, for example, user ID, user name, age, gender, number of times of posting, product genre handled, and the like. Viewer attributes, viewer trends, and excitement trends are analysis results by the analysis server 600 .

FIG. 5 is a diagram showing an example of the contents of the third reference information 733. As shown in FIG. The third reference information 733 is, for example, information in which the second related word is associated with the time information. A second related word is a word included in the message information. The time information is, for example, information indicating the timing at which the second related word appears in the live moving image.

The processing unit 750 includes, for example, a generating unit 751 and a reading unit 752 . Some or all of these components are realized by executing a program (software) stored in the storage unit 730 by a processor such as a CPU, for example. In addition, some or all of the functions of these components may be realized by hardware such as LSI, ASIC, FPGA (including circuitry), or by cooperation of software and hardware. may be implemented.

Generation unit 751 generates first reference information 731 , second reference information 732 , and third reference information 733 based on information received from analysis server 600 using communication unit 710 . Generation unit 751 updates storage unit 730 according to the type of information included in the analysis result received using communication unit 710 .

For example, the generation unit 751 extracts the first related word from text data received from the analysis server 600 (text data converted from voice data). The generation unit 751 generates the first reference information 731 by associating the extracted first related word with time information indicating the timing at which the first related word appears in the live video. For example, the generation unit 751 extracts, as first related words, nouns and words registered in a dictionary prepared in advance from among the words included in the text data received from the analysis server 600 .

In addition, the generation unit 751 generates the second reference information 732 by associating the live ID received from the analysis server 600 with the set channel, meta information, poster attribute, viewer attribute, viewer trend, and excitement trend. .

The generation unit 751 also extracts second related words from the message information received from the analysis server 600 . The generation unit 751 generates the third reference information 733 by associating the extracted second related word with time information indicating the timing at which the second related word appears in the live video. For example, the generation unit 751 extracts, as second related words, nouns and words registered in a dictionary prepared in advance from among the words included in the message information received from the analysis server 600 .

When the reading unit 752 receives a reference information transmission request from the recommendation distribution server 800 using the communication unit 710 , the reading unit 752 reads the designated reference information from the storage unit 730 and transmits it to the recommendation distribution server 800 . Note that the reading unit 752 may read only a designated part of the reference information stored in the storage unit 730 and transmit it to the recommendation distribution server 800 .

(Recommendation distribution server 800)
The recommendation distribution server 800 includes a communication unit 810, a storage unit 830, and a processing unit 850, for example. The communication unit 810 includes, for example, a communication interface such as NIC. The storage unit 830 is, for example, a RAM, a ROM, a flash memory such as an SSD, an HDD, or the like. The storage unit 830 stores, for example, first reference information 831, second reference information 832, third reference information 833, user information 834, ranking information 835, and the like. The storage unit 830 may be an external storage device such as a NAS that can be accessed by the recommendation distribution server 800 via a network.

FIG. 6 is a diagram showing an example of the contents of the user information 834. As shown in FIG. The user information 834 is information about the user, and includes information preset by the user, history information of the user using the content provided by the information providing server 400, and the like. The user information 834 is, for example, information in which a user ID is associated with attribute information, preference information, access history, search query history, and the like. A user ID is information for identifying each user. The attribute information is information about attributes of the user indicated by the user ID, and includes, for example, user name, age, and gender. The preference information is information about the user's preference indicated by the user ID, and is derived based on, for example, the tendency of the preference set by the user (for example, hobbies, interests, etc.) and the browsing history of the user's content. It includes the tendency of preference. The access history is the past content access history by the user indicated by the user ID. The search query history is a history of past search queries used by the user indicated by the user ID.

FIG. 7 is a diagram showing an example of the contents of the ranking information 835. As shown in FIG. The ranking information 835 is, for example, information in which a ranking is associated with an overall ranking, a ranking by category, a ranking by user attribute, a rapid rise ranking, and the like. All live videos are eligible for the overall ranking and soaring ranking. The ranking by category is for live videos classified according to the category of the content of the live videos (for example, sports, housework, beauty, etc.). The ranking by user attribute targets live videos classified according to the attribute of the distributor and the attribute of the audience.

The processing unit 850 includes, for example, an update unit 851, an information acquisition unit 852, a derivation unit 853, a determination unit 854, and a distribution unit 855. Some or all of these components are realized by executing a program (software) stored in the storage unit 830 by a processor such as a CPU, for example. In addition, some or all of the functions of these components may be realized by hardware such as LSI, ASIC, FPGA (including circuitry), or by cooperation of software and hardware. may be implemented.

The update unit 851 transmits a reference information transmission request to the index server 700 at a predetermined timing. The update unit 851 updates the first reference information 731, the second reference information 732, and the third reference information 733 stored in the storage unit 830 based on the information received from the index server 700. FIG.

The information acquisition unit 852 uses the communication unit 810, for example, to acquire request information or the like transmitted from the terminal device 500 of the service user C (or information based on the request information, the same shall apply hereinafter). Note that the information acquisition unit 852 may receive request information or the like from the information providing server 400 or directly from the terminal device 500 . The information acquisition unit 852 updates the user information 834 stored in the storage unit 830 based on the acquired request information. The information acquisition unit 852 may store the request information received from the information providing server 400 as it is in the storage unit 830, or store the information generated based on the request information received from the information providing server 400 in the storage unit 830. You may Note that the information acquisition unit 852 may acquire request information based on information received from an external server other than the information providing server 400 .

For example, the information acquisition unit 852 acquires information preset by the service user C as request information. The information set in advance by the service user C includes, for example, attribute information of the service user C (for example, age, gender, area of residence, school attended, occupational genre, etc.) and preference information (for example, hobbies, interests, etc.). etc.) etc. The information set in advance by the service user C may be based on registration information registered in advance in the information providing server 400 in association with the user ID included in the request information. It may be based on registration information registered in advance in storage unit 830 in association with the included user ID.

For example, the information acquisition unit 852 acquires information about the content accessed by the service user C as request information. For example, based on the access history received from the information providing server 400, information on the content accessed by the service user C is acquired. The content accessed by the service user C includes, for example, content accessed within a predetermined time from the current time, the latest content among the accessed content, the content being accessed, the content being accessed and not yet opened in the terminal device 500 This includes content that is not

For example, the information acquisition unit 852 acquires information about preferences of the service user C as request information. For example, the information acquisition unit 852 acquires information about preferences of service user C based on registration information registered in advance by service user C in association with the user ID included in the request information. In addition, the information acquisition unit 852 analyzes the content viewing history of service user C in association with the user ID included in the request information, derives the tendency of service user C's tastes, and extracts the tastes of service user C. You may obtain information about The browsing history of the content of the service user C includes at least one of the information regarding the content that the service user C has accessed in the past and the information regarding the search queries that the service user C has used in the past.

The request information is not limited to the information described above, and may include, for example, information on search query history, information on past online shopping purchase history, information on applications installed on the user's own terminal device, and the like.

Derivation unit 853 derives a matching score based on information stored in storage unit 830 . The matching score includes a first matching score and a second matching score. The first matching score is information used to determine a live video that matches service user C, and is information about the content of the live video that changes according to the elapsed time. The second matching score is information used to determine a live video that matches the live video trend, and is information about the live video trend that changes according to the elapsed time. A trend represents the trend of the latest fad in a live video, and is, for example, a fad represented based on the number of viewers, the number of conversions, the number of messages, and the like.

For example, the derivation unit 853 refers to the first reference information 831 and derives the first matching score for each first related word by multiplying the first related word by a predetermined weight. For example, the predetermined weight increases as the time information associated in the first reference information 831 approaches the current time. Also, when the first reference information 831 of a given live video includes a plurality of common first related words, the derivation unit 853 multiplies each of the common first related words by a predetermined weight, and obtains a score in terms of the number of words. Normalization may be used to derive a first matching score for common first related words. Common first related words include, for example, words with matching letters, words with matching meanings, words with related meanings, and the like. Note that the derivation unit 853 may refer to the third reference information 831 and derive the first matching score for the second related word (or the common second related word) by performing the same processing as described above. good.

The deriving unit 853 also refers to the second reference information 832 to derive the second matching score. For example, the derivation unit 853 derives the score per unit time by referring to the second reference information 832 and multiplying the number of viewers per unit time by a predetermined weight. For example, the predetermined weight increases as the time information associated in the second reference information 832 approaches the current time. For example, the derivation unit 853 derives the total score per unit time as the second matching score. Similarly, the deriving unit 853 may derive a second matching score based on the number of conversions, a second matching score based on the number of messages, and the like.

Further, the derivation unit 853 may derive the second matching score by adding a predetermined value based on the viewer behavior of the second reference information 832 when the viewer behavior satisfies the first condition. The first condition includes, for example, that the audience rating is equal to or higher than a threshold, and that the ratio of the number of people who have made a purchase to the number of viewers is equal to or higher than a threshold. Note that the derivation unit 853 may derive the second matching score by adding a predetermined value that increases stepwise as the audience rating increases.

Also, the deriving unit 853 creates ranking information 835 indicating the ranking of live moving images in predetermined items based on the information stored in the storage unit 830 . For example, the deriving unit 853 derives the overall ranking, the ranking by category, the ranking by user attribute, the sudden increase ranking, etc. based on the reference information corresponding to the past target period within a predetermined time from the current time.

The deriving unit 853 may derive each rank for each item, or may derive each rank by comprehensively evaluating a plurality of items. The items measure the popularity, trends, advertising effectiveness, etc. of live videos, and include, for example, viewing rate, conversion rate, message rate, surge rate, product purchase rate, and the like. The audience rating is, for example, the average number of viewers of a live video within a predetermined period of time, or the ratio of the number of viewers of one live video to the number of viewers of all live videos being distributed. The conversion rate is, for example, the average number of conversions of live videos within a predetermined period of time, or the ratio of the number of conversions of one live video to the number of conversions of all live videos being distributed. The message rate is, for example, the average number of messages included in live videos within a predetermined period of time, or the ratio of the number of messages included in one live video to the number of messages included in all live videos being distributed. The rapid increase rate is the rate at which the number of viewers and the number of messages per unit time increased sharply compared to the average number of viewers and the number of messages for one live video. The product purchase rate is the number of purchases of the target product or the like purchased during distribution of the live video, the number of purchases per unit time, the maximum number of purchases per unit time, and the like.

Further, the deriving unit 853 may change the ranking based on the remaining time of the distribution period of the live video. For example, when the remaining time of the live video is five minutes or less, the deriving unit 853 removes the target live video from the ranking, or lowers the ranking of the target live video by 10 places or more.

Further, the deriving unit 853 may exclude information about users who viewed live videos via recommendations distributed by the recommendation distribution server 800 when deriving matching scores and rankings. In addition, the deriving unit 853 may derive matching scores and rankings separately for counting users who have viewed the live video via the recommendations distributed by the recommendation distribution server 800 and counting for other users. good.

The determination unit 854 refers to the information stored in the storage unit 830, and based on the request information acquired by the information acquisition unit 852, selects a push video to be presented to the service user C from among the live videos being distributed. to decide. For example, when the determination unit 854 determines the push video based on the information about service user C included in the request information, the recommendation distribution server 800 provides service user C who likes gadgets with the latest smartphone-related videos. You can push live videos about your products. In addition, when the determination unit 854 determines the push video based on the information about the content of the provided information included in the request information, the recommendation distribution server 800 browses the content of articles about the latest gadgets and the content about the products of the latest gadgets. It is possible to push live videos regarding the latest smartphone-related products to service user C who is using the service. By doing so, it is possible to efficiently recommend live moving images.

In addition, the determination unit 854 determines a plurality of candidates, and selects, from among the plurality of candidates, the content of the most recent live video that matches the service user C or the trend of the live video. may be determined to For example, the determination unit 854 refers to the matching score derived by the derivation unit 853 in addition to these pieces of information (reference information and request information) to determine push videos from the live videos being distributed. Specifically, the determining unit 854 determines a plurality of candidates, and determines a live video with a high matching score from among the plurality of candidates as a push video. A plurality of candidates are determined, for example, using any method for determining push animations. Live videos with a high matching score include, for example, live videos with a first matching score and a second matching score equal to or greater than a threshold, live videos with the highest total value of the first matching score and the second matching score, and the like. . The determination unit 854 can select the most recent live video that matches the content of the service user C by using the first matching score. The determining unit 854 can select live video that matches the trend of the live video by using the second matching score. By doing so, it is possible to select a live moving image with a higher advertising effect from among the live moving images suitable for the service user C, and push the selected live moving image to the service user C.

In addition to these pieces of information, the determination unit 854 may refer to the ranking information 835 to determine the push video from among the live videos being distributed. For example, the determining unit 854 may determine a plurality of push movie candidates, and determine a live movie with a high ranking among the plurality of push movie candidates as a push movie. Live videos with high rankings include, for example, live videos with high rankings included in the ranking information 835, live videos with high overall evaluations in the rankings included in the ranking information 835, and the like.

For example, the determination unit 854 selects the live video related to the content accessed by the service user C based on the “information on the content accessed by the service user C” acquired by the information acquisition unit 852 as a push video. to decide. For example, if the service user C has accessed the contents of a candy maker, the determination unit 854 selects live videos related to sweets (for example, live videos introducing sweets products, Live videos to be introduced) are decided as push videos. By doing so, it is possible to push to the service user C the live video that matches the content that the service user C has just accessed or the content that the service user C is currently accessing.

Note that the determining unit 854 refers to the first matching score derived by the deriving unit 853, and among live videos related to sweets, a live video having a high first matching score for words related to sweets is pushed. You can decide on a video. By doing so, it is possible to determine the push video according to the trend of the distributed live video. As a result, for example, a live video that was about sweets until five minutes ago but is currently about something other than sweets can be excluded from the push videos.

Further, the determination unit 854 may refer to the ranking information 835 derived by the derivation unit 853 and determine a live video with a high rank among live videos related to sweets as a push video. By doing so, it is possible to determine a live video with a high degree of attention among distributed live videos to be a push video.

Also, based on the information about service user C's preferences, the determination unit 854 determines a live video that matches service user's C preferences as a push video. For example, if the information about service user C's preferences includes soccer, baseball, etc., the determination unit 854 determines live videos related to soccer and baseball, and live videos related to other sports as push videos. do. By doing so, it is possible to push to the service user C a live moving image that matches the taste of the service user C. In addition, even if there is too much or little information about the accessed content like service user C who accessed the top page for search, it is possible to determine a push video suitable for service user C. can be done.

Note that the determining unit 854 may refer to the first matching score derived by the deriving unit 853 and determine a live video with a high first matching score for words related to soccer or baseball as a push video. Further, the determination unit 854 may refer to the ranking information 835 derived by the derivation unit 853 and determine a live video with a high ranking among live videos related to soccer or baseball as a push video.

Also, the determination unit 854 may determine a push video based on the attribute of the service user. For example, if the service user C is a woman in her twenties, the determining unit 854 refers to an attribute dictionary prepared in advance to obtain keywords that women in their twenties are likely to be interested in, and matches the keywords thus obtained. A live video containing a related word or a live video containing message information that matches the acquired keyword is determined as a push video. Further, the determining unit 854 determines, as a push video, a live video posted by a poster with attributes matching the attributes of service user C (same age, same region, same school, same occupational genre, etc.). may By doing so, it is possible to push to the service user C a live video that matches the attributes of the service user C. Note that the determining unit 854 refers to the first matching score derived by the deriving unit 853, refers to an attribute dictionary prepared in advance, and refers to the first matching score of words related to keywords that women in their twenties are likely to be interested in. A live video with a high matching score may be determined as a push video.

Also, the determination unit 854 may refer to the related word included in the first reference information 831 and the message information included in the third reference information 833 to determine the push video. For example, if service user C's hobby is yoga, the determination unit 854 selects a live video containing a related word related to yoga among the related words included in the first reference information 831, or a related word included in the third reference information 833. Among the message information received, a live video including message information related to yoga is determined as a push video. Note that service user C's hobbies are included in the preference information of the user information 834 . By using related words and message information in this way, a live moving image that matches the service user C can be easily selected. Note that the determining unit 854 may refer to the first matching score derived by the deriving unit 853 and determine a live video with a high first matching score for words related to yoga as a push video.

Also, the determination unit 854 may determine a live video whose viewer trend satisfies the first condition as a push video based on the viewer trend of the second reference information. The first condition includes, for example, that the audience rating is equal to or higher than a threshold, and that the ratio of the number of people who have made a purchase to the number of viewers is equal to or higher than a threshold. By doing so, it is possible to push live videos that are popular with viewers to the service user C. Note that the determination unit 854 may determine a live video that satisfies the first condition for viewer trends among viewer trends that match the attribute of service user C as a push video. By doing so, it is possible to push to service user C a live moving image that is highly popular among viewers whose attributes match those of service user C.

Further, the determination unit 854 may determine, as the push video, a live video whose trend of excitement satisfies the second condition based on the trend of excitement of the second reference information. The second condition includes, for example, that the degree of excitement is equal to or greater than a threshold, that the timing at which the degree of excitement becomes equal to or greater than the threshold is within a predetermined time from the current time, and the duration of the period during which the degree of excitement is equal to or greater than the threshold. It includes being equal to or higher than a threshold. By doing so, it is possible to push the live action that attracts a lot of attention to the service user C. Note that the determining unit 854 may determine a live video that satisfies the second condition in terms of the trend of excitement among viewers whose attributes match the attribute of the service user C, among the trend of excitement, as the push video. By doing so, it is possible to push to service user C a live moving image that is popular among viewers whose attributes match those of service user C.

The distribution unit 855 distributes information about the push video determined by the determination unit 854 to the service user C. FIG. The information about the push animation includes a URL for accessing the push animation, a still image or animation that accesses the push image by being clicked, a push animation, and the like. Note that the distribution unit 855 transmits, for example, information about the push video to the information providing server 400 . The information providing server 400 embeds the information related to the push video received from the distribution unit 855 into part of the content to be provided to the service user C, and transmits the content to the terminal device 500 of the service user C. FIG. By doing so, the service user C can access the live distribution server 100 and view the push video.

In addition, when distributing information about push videos to service user C, the distributing unit 855, if a live video with a high matching score is determined as a push video, indicates that fact (including the content of the matching score with a high score). may be delivered. In addition, when a live video with a high ranking is determined as a push video, the distribution unit 855 may distribute information to that effect (including the content of the ranking information of the high ranking) to the service user. By doing so, the willingness of the service user C to browse can be aroused.

[Sequence Diagram]
Next, referring to FIG. 8, processing executed in the information distribution system 1 will be described. FIG. 8 is a sequence diagram showing processing in the information distribution system 1 according to this embodiment.

It is assumed that live contributor A has transmitted a distribution request to live distribution server 100 and set a distribution channel in advance. The terminal device 200 captures a live video (S1), and transmits the captured live video to the live distribution server 100 (S2). The live distribution server 100 transmits the information on the live video received from the terminal device 200 to the analysis server 600 (S3), and distributes the information on the distribution channel (S4).

The analysis server 600 analyzes the received information about the live moving image (S5), and transmits the analysis result to the index server 700 (S6). The index server 700 generates reference information (first reference information 731, second reference information 732, and third reference information 733) based on the analysis results received from the analysis server 600 (S7). For example, when the reference information update timing has arrived (S8), the recommendation distribution server 800 transmits a reference information transmission request to the index server 700 (S9). The index server 700 reads reference information from the storage unit 730 (S10) and transmits it to the recommendation distribution server 800 (S11). The recommendation distribution server 800 stores the received reference information in the storage unit 830 (S12).

Next, it is assumed that live viewer B uses the terminal device 300 to give an instruction to view the live video being distributed (S31). The terminal device 300 transmits to the live distribution server 100 a distribution request for the live moving image specified by the operation of the live viewer B (S32). The live distribution server 100 reads the moving image data and the audio data of the designated live moving image from the distribution folder 131 (S33), and transmits the read data to the terminal device 300 (S34).

Next, it is assumed that live viewer B uses the terminal device 300 to post a comment on the live video being viewed (S51). The terminal device 300 transmits the posted comment to the live distribution server 100 (S52). After temporarily storing the received comment in the message folder 133, the live distribution server 100 reflects it in the live video in the distribution folder 131 (S53). Then, the live distribution server 100 distributes the live moving image in the distribution folder 131 through the distribution channel, whereby the live moving image reflecting the comments is distributed from the distribution channel (S54). In parallel with this, the live distribution server 100 transmits information about the live video in which the comments are reflected to the analysis server 600 (S55). The analysis server 600 analyzes the information about the live video reflecting the received comment (S56), and transmits it to the index server 700 (S57).

On the other hand, it is assumed that service user C uses terminal device 500 to access content provided by information providing server 400 (S71). The information providing server 400 acquires the accessed content (S72) and transmits information on the accessed content to the recommendation distribution server 800 (S73). The recommendation distribution server 800 acquires the request information, refers to the reference information, and determines the push video (S74). The recommendation distribution server 800 transmits information about the determined push video to the information providing server 400 (S75). The information providing server 400 embeds the received information about the push animation in a part of the accessed content (S76), and transmits the content embedded with the information about the push animation to the terminal device 500 of the service user C (S77). ).

[Flowchart of live distribution server]
Next, with reference to FIG. 9, processing executed in live distribution server 100 will be described. FIG. 9 is a flow chart showing an example of processing executed in the live distribution server 100. As shown in FIG.

For example, the setting unit 151 determines whether or not a distribution request has been received from the live poster A using the communication unit 110 (S101). When receiving the distribution request, the setting unit 151 sets a distribution channel for distributing the live video posted by the live contributor A (S103). Next, the data management unit 152 receives information about the posted live video from the terminal device 200 and stores it in the distribution folder 131 (S105). At this time, the data management unit 152 creates meta information about the posted live video and stores it in the distribution folder 131 . The data management unit 152 transmits the analysis target data read from the distribution folder 131 to the analysis server 600 (S107). Also, the data management unit 152 moves the information read from the distribution folder 131 to the archive folder 132 (S109).

Next, the data management unit 152 determines whether or not the next live video has been received from the terminal device 200 (S111). When the next live moving image is received from the terminal device 200, the data management unit 152 stores information regarding the received live moving image in the distribution folder 131 (S113). At this time, the data management unit 152 creates meta information about the posted live video and stores it in the distribution folder 131 .

Next, the data management unit 152 determines whether or not a live video viewing request has been received from the terminal device 300 of the live viewer B (S115). When receiving a live video viewing request from the terminal device 300, the data management unit 152 transmits the information read from the distribution folder 131 to the terminal device 300 (S117).

Next, the data management unit 152 determines whether message information has been received from the terminal device 300 (S119). When receiving message information from the terminal device 300, the data management unit 152 stores the received message information in the message folder 133 (S121).

Next, the data management unit 152 determines whether purchase information has been received from the terminal device 300 (S123). The purchase information includes, for example, information on products purchased via the live video viewed by the live viewer B, and the like. When the purchase information is received from the terminal device 300, the data management unit 152 includes the received purchase information in the meta information of the live video and stores it in the distribution folder 131 (S125).

Then, the data management unit 152 transmits the analysis target data read from the distribution folder 131 and the message folder 133 to the analysis server 600 (S127). The data management unit 152 moves the information read from the distribution folder 131 and the information read from the message folder 133 to the archive folder 132 (S129).

Next, the data management unit 152 determines whether or not the distribution of the live video by the live contributor A has ended (S131). If the distribution of the live video by the live contributor A has not ended, the data management unit 152 returns to S111 and repeats the process. When the distribution of the live video by the live contributor A ends, the data management unit 152 ends the process.

[Index server flow chart]
Next, referring to FIG. 10, processing executed in the index server 700 will be described. FIG. 10 is a flow chart showing an example of processing executed in the index server 700. As shown in FIG. Note that the index server 700 executes the processing described below for each set channel.

The generation unit 751 determines whether or not an analysis result or the like has been received from the analysis server 600 (S201). When the analysis result or the like is received from the analysis server 600, the generation unit 751 determines whether text data is included in the received analysis result (S203). When text data is included in the received analysis result, the generation unit 751 extracts the first related word from the text data, associates the time information indicating the timing at which the first related word appears, and generates the first reference word. The information 731 is updated (S205).

Next, the generation unit 751 determines whether message information is included in the received analysis result (S207). If message information is included in the received analysis result, the generation unit 751 extracts the second related word from the message information, associates the time information indicating the timing at which the second related word appears, and generates the third reference word. The information 733 is updated (S209).

Next, the generation unit 751 determines whether or not the received analysis results include other analysis results (S211). If other analysis results are included in the received analysis results, the generator 751 updates the second reference information 732 based on the other analysis results (S213).

[Flowchart of recommendation distribution server]
Next, processing executed in the recommendation distribution server 800 will be described with reference to FIGS. 11 and 12 are flowcharts showing an example of processing executed in the recommendation distribution server 800. FIG. Note that the recommendation distribution server 800 executes the processing described below for each service user.

For example, based on the information received from the information providing server 400, the information obtaining unit 852 determines whether the information providing server 400 has received request information from the terminal device 500 of the service user C (S301). When the information providing server 400 receives the request information, the information acquisition unit 852 acquires the request information (S303).

Next, the information acquisition unit 852 determines whether or not information related to the content accessed by the service user C has been acquired as request information (S305). When the information about the content accessed by service user C is acquired, the determining unit 854 includes the live video related to the content accessed by service user C in the push video candidates (S307).

Next, the information acquisition unit 852 determines whether or not the information regarding the preferences of the service user C has been acquired as the request information (S309). If the information about the preferences of service user C has been acquired, the determining unit 854 includes live videos that match the preferences of service user C in the push video candidates (S311).

Next, the information acquisition unit 852 determines whether or not attribute information of the service user has been acquired as request information (S313). When the attribute information of the service user has been acquired, the determining unit 854 includes the live video that matches the attribute of the service user in the push video candidates (S315).

Note that the determination unit 854 is not limited to this, and may determine push video candidates using all of the acquired request information. may be determined.

Next, the determination unit 854 determines whether or not the push movie candidates include a plurality of live movies (S317). When the push movie candidates include a plurality of live movies, the determination unit 854 determines whether matching scores have been derived for all the live movies included in the push movie candidates (S319). When matching scores have been derived for all live videos included in the push video candidates, the determination unit 854 refers to the matching scores derived by the derivation unit 853, and selects live videos with matching scores equal to or lower than the threshold as push video candidates. (S321).

Next, the determination unit 854 refers to the ranking information 835 and determines whether or not there is a live video ranked high (for example, within the top 5) in any of the rankings among the push video candidates ( S323). When a live video ranked high in any ranking is included in the push video candidates, the determination unit 854 determines the live video ranked high as a push video (S325). Then, the distribution unit 855 transmits information on the push video determined by the determination unit 854 to the service user C (S327).

According to the embodiments described above, an analysis unit that analyzes information about a live video being distributed, a generation unit that generates reference information based on the analysis result of the analysis unit, and a service user who has transmitted request information. an acquisition unit that acquires the request information from an information providing server that transmits the provided information to the terminal device; referring to the reference information generated by the generation unit; and based on the request information that is acquired by the acquisition unit, a determination unit that determines a live video to be presented to the service user from among the live videos that are being distributed; and a distribution unit that distributes information about the live video determined by the determination unit to the service user. By having it, a live video can be efficiently recommended using a live distribution service.

As described above, the mode for carrying out the present invention has been described using the embodiments, but the present invention is not limited to such embodiments at all, and various modifications and replacements can be made without departing from the scope of the present invention. can be added.

For example, the information providing server 400 and the recommendation distribution server 800 may be one server, or may be one server including other servers.

For example, when determining a push movie, a live movie with a high overall evaluation value may be determined as a push movie without determining a candidate for a push movie.

In the above description, an example in which service user C is a user who has not viewed live video has been described, but the present invention is not limited to this. Service user C may be a user viewing a live video. For example, assume that service user C transmits request information including a query while viewing live video J. In this case, for example, the recommendation distribution server 800 determines a push movie corresponding to the query, and the information providing server 400 sends the content embedded with the push movie to the search result based on the query to the terminal device 500 of the service user C. Send. The terminal device 500 displays the content received from the information providing server 400 on a part of the screen while displaying the live video J, for example. By doing so, it is possible to push to the service user C a live video K different from the live video J currently being viewed from among the live videos currently being distributed.

In the above description, an example has been described in which the request information is information requesting transmission of information other than live video, but the present invention is not limited to this. The request information may request transmission of information about the live video. For example, assume that service user C specifies a genre and a keyword of a live video, and transmits request information requesting transmission of information on the live video related to the specified genre and keyword. Note that the request information is not limited to this, and may be, for example, information about the service user C, as described above. In this case, for example, the recommendation distribution server 800 determines a push movie corresponding to the specified genre or keyword, and transmits content including information about the push movie to the terminal device 500 of the service user C. FIG. The terminal device 500 displays, for example, the content received from the recommendation distribution server 800 on the screen. By doing so, it is possible to efficiently recommend the live moving image that is currently being delivered to the service user C.

DESCRIPTION OF SYMBOLS 1... Information distribution system 100... Live distribution server 110... Communication part 130... Storage part 150... Processing part 151... Setting part 152... Data management part 200... Terminal device 300... Terminal device 400... Information providing server 500 Terminal device 600 Analysis server 610 Communication unit 630 Storage unit 650 Analysis unit 700 Index server 710 Communication unit 730 Storage unit 750 Processing unit 751 Generating unit 752 Reading unit 800 Recommendation distribution server 810 Communication unit 830 Storage unit 850 Processing unit 851 Update unit 852 Information acquisition unit 853 Derivation unit 854 Determination unit , 855 ... distribution unit 855

Claims (13)

an analysis unit that analyzes information about the live video being distributed;
a generation unit that generates reference information based on the analysis result of the analysis unit;
an acquisition unit that acquires web page request information transmitted from a terminal device of a service user;
A determination unit that refers to the reference information generated by the generation unit and determines a live video to be presented to the service user from among the live videos being distributed based on the request information acquired by the acquisition unit. When,
a distribution unit that distributes information about the live video determined by the determination unit to the service user;
with
The request information is information requesting transmission of information other than the live video,
Information delivery system. The distribution unit distributes the live video determined by the determination unit to the service user in a form embedded in the web page requested by the request information.
The information distribution system according to claim 1. The generating unit generates, as the reference information, information in which time information indicating a timing corresponding to the analysis result on the time axis in the live video being distributed is associated with the analysis result,
further comprising a first derivation unit that uses the reference information to derive information about the content of the live video being distributed that changes according to the elapsed time;
The determination unit determines a live video to be presented to the service user based on the information derived by the first derivation unit.
3. The information distribution system according to claim 1 or 2. The first derivation unit uses the reference information to derive information about the trend of the live video being distributed that changes according to the elapsed time.
The information distribution system according to claim 3. The acquisition unit acquires information about content of provided information according to the request information provided by the information providing server,
The determining unit determines a live video related to the content of the provided information as a live video to be presented to the service user, based on information about the content of the provided information.
The information distribution system according to any one of claims 1 to 4. The acquisition unit acquires information about preferences of the service user,
The determination unit determines a live video that matches the service user's preference as a live video to be presented to the service user, based on the information about the service user's preference acquired by the acquisition unit.
The information distribution system according to any one of claims 1 to 5. further comprising a second derivation unit that derives the ranking of the live video in a predetermined item based on the information about the live video being distributed;
The determination unit determines the live video to be presented to the service user based on the ranking derived by the second derivation unit.
The information delivery system according to any one of claims 1 to 6. The analysis unit analyzes the audio data included in the live video being distributed to convert the audio data into text data,
The generation unit extracts a first related word related to the live video being distributed from the text data, and includes the extracted first related word in the reference information.
The information distribution system according to any one of claims 1 to 7. The analysis unit acquires post information corresponding to the live video being distributed,
The generation unit extracts second related words related to the live video being distributed from the posted information, and includes the extracted second related words in the reference information.
The information distribution system according to any one of claims 1 to 8. The determination unit refers to the reference information generated by the generation unit, determines a plurality of candidates from the live video being distributed based on the request information acquired by the acquisition unit, and determines determining a live video to be presented to the service user from among the plurality of candidates whose content of the most recent live video matches the service user;
The information delivery system according to any one of claims 1 to 9. The analysis unit acquires trends of viewers of the live video as the analysis result of information related to the live video being distributed,
The generation unit includes the acquired trends of the viewer in the reference information,
The information distribution system according to any one of claims 1 to 10. The analysis unit obtains, as the analysis result of the information related to the live video being distributed, a trend of excitement of the live video,
The generation unit includes the acquired trend of the excitement in the reference information,
The information distribution system according to any one of claims 1 to 11. one or more computers
Analyze information about the live video being distributed,
generate reference information based on the analysis results,
Acquire web page request information sent from the terminal device of the service user,
referring to the reference information and determining a live video to be presented to the service user from among the live videos being distributed based on the request information;
An information distribution method for distributing information about a live video to be presented to the determined service user to the service user,
The request information is information requesting transmission of information other than the live video,
Information delivery method.

Images