News information analysis equipment, news information analysis program, and news information analysis method

Abstract

To allow a user who has highly frequently accesses news information to efficiently browse articles in a wide range of genres.SOLUTION: News information analysis equipment 1 comprises: a news information acquisition unit 110 that acquires a plurality of pieces of news information from a communication network, assigns news identification information to the news information, and stores it; an output unit 111 that provides the news information to users for viewing; a user information management unit 112 that tallies up the number of times the news information provided for viewing has been viewed for each user; and an output control unit 114 that calculates, for a user whose number of times the user has viewed the news information is equal to or more than a preset threshold, a similarity between news information that has been viewed by the user and news information that has not been viewed by the user, and controls the output unit 111 in such a manner that the news information that has not been viewed by the user and whose similarity is equal to or less than a preset prescribed value, is recommended to the user.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a news information analysis program and a news information analysis method that enable a user who frequently accesses multiple news information distributed on the Internet to be displayed news information that has a low similarity to the news information viewed by the user. [Background technology]

[0002] Due to the vast amount of news information distributed on the Internet, users need to be able to efficiently browse news that matches their interests. For this reason, technologies for organizing news information are currently being developed. For example, Patent Document 1 discloses a technology in which an information processing device automatically collects and presents useful information of interest to a user via the Internet by extracting and presenting recommended articles based on keywords of interest and friends' ratings. Such devices always present articles that are of interest to the user, but for heavy users who frequently read articles, the devices always present similar articles, which can lead to a rut and are not suitable for users looking for articles across a wide range of genres. In such a case, it may lead to a further decrease in the number of times the service is used. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2012-216072 Summary of the Invention [Problem to be solved by the invention]

[0004] Because the amount of news information distributed on the Internet is enormous, there is a demand for a news information analysis device, a news information analysis program, and a news information analysis method that enable even users who frequently access news information to efficiently browse articles across a wide range of genres.

[0005] The present invention has been made in consideration of such problems, and aims to provide a news information analysis device, news information analysis program, and news information analysis method that acquires multiple pieces of news information from a communication network and provides the news information to users for viewing, and that can calculate the similarity between the news information viewed by a user (hereinafter, unless otherwise specified, referred to as a "user who frequently accesses news information") whose number of times the news information has been viewed is equal to or exceeds a predetermined threshold value, and recommend to the user news information that the user has not viewed and whose similarity is equal to or below a predetermined value. [Means for solving the problem]

[0006] (1) A news information analysis device according to one aspect of the present invention (e.g., news information analysis device 1 described later) includes a news information acquisition unit (e.g., news information acquisition unit 110 described later) that acquires multiple pieces of news information including text information from a communication network, assigns news identification information to each piece of news information, and stores the information; an output unit (e.g., output unit 111 described later) that provides the news information stored by the news information acquisition unit for viewing by users; and a user information management unit (e.g., The news information management system includes a user information management unit 112 described below), a similarity calculation unit (e.g., a similarity calculation unit 113 described below) that calculates the similarity between the news information viewed by the user and assigned with the identification information, and other news information not viewed by the user, for the user whose number of views of the news information is equal to or greater than a predetermined threshold, and an output control unit (e.g., an output control unit 114 described below) that controls the output unit to recommend to the user the other news information not viewed by the user, where the similarity is equal to or less than a predetermined value set in advance.

[0007] According to (1) above, even a user who frequently accesses news information can efficiently view articles in a wide range of genres that the user has not yet viewed.

[0008] (2) In the news information analysis device described in (1) above (for example, the news information analysis device 1 described below), the number of times the news information is viewed may be any one of the number of news information items with different news identification information accessed by each user, the number of times the news information items accessed by each user are accessed, or the number of times a news information providing site that provides the news information for viewing by each user is accessed.

[0009] According to (2) above, it becomes possible to efficiently view articles in a wide range of genres that the user has not yet viewed, based on the frequency of access to news information that is compiled from various perspectives.

[0010] (3) In the news information analysis device described in (1) or (2) above (e.g., the news information analysis device 1 described below), the similarity calculation unit (e.g., the similarity calculation unit 113 described below) may calculate the similarity based on the degree of similarity between the text information included in the news information.

[0011] According to (3) above, even a user who frequently accesses news information can efficiently view articles in a wide range of genres that the user has not yet viewed.

[0012] (4) In the news information analysis device described in (1) or (2) above (e.g., the news information analysis device 1 described below), the output control unit (e.g., the output control unit 114 described below) may give priority to newer news information acquired by the news information acquisition unit when recommending news information to the user.

[0013] According to (4) above, it becomes possible to efficiently browse fresh articles in a wide range of genres.

[0014] (5) In the news information analysis device described in (1) or (2) above (e.g., the news information analysis device 1 described later), the output control unit (e.g., the output control unit 114 described later) may preferentially recommend to the user the news information having the news identification information whose number of views by a user other than the user is equal to or greater than a preset value, for other news information whose similarity is equal to or less than a preset value and whose number of views by a user other than the user is equal to or greater than a preset threshold value and whose number of views by the news information is equal to or greater than a preset threshold value.

[0015] According to (5) above, it becomes possible to efficiently browse fresh articles in a wide range of genres.

[0016] (6) A news information analysis program according to one aspect of the present invention includes a computer device connected to a communication network (e.g., communication network 60 described below), a news information acquisition unit (e.g., news information acquisition unit 110 described below) that acquires multiple pieces of news information including text information from the communication network, assigns news identification information to each piece of news information, and stores the information, an output unit (e.g., output unit 111 described below) that provides the news information stored by the news information acquisition unit for viewing by users, and a program that identifies users who view the news information provided for viewing by the output unit and tallies the number of times the news information is viewed for each user. the user information management unit (e.g., user information management unit 112 described later) that calculates the similarity between the news information to which the identification information has been assigned and which the user has viewed, and other news information that the user has not viewed, for the user whose number of views of the news information is equal to or greater than a preset threshold; and an output control unit (e.g., output control unit 114 described later) that controls the output unit to recommend to the user other news information that the user has not viewed, and whose similarity is equal to or less than a preset predetermined value.

[0017] According to (6) above, the same effect as the news information analysis device according to (1) above can be achieved.

[0018] (7) The information analysis method of the present invention is a news information analysis method executed by a computer, and includes a news information acquisition step of acquiring a plurality of news information including text information from a communication network, assigning news identification information to each piece of news information, and storing the news information; an output unit (e.g., output unit 111 described below) that provides the news information stored in the news information acquisition step to be viewed by users; a user information management step of identifying users who view the news information provided for viewing in the output step and tallying up the number of times the news information has been viewed for each user; a similarity calculation step of calculating, for a user whose number of times the news information has been viewed is equal to or greater than a predetermined threshold, a similarity between the news information assigned with the identification information and viewed by the user, and other news information not viewed by the user; and an output control step of controlling the output step to recommend to the user other news information not viewed by the user, the similarity of which is equal to or less than a predetermined value set in advance.

[0019] According to the method (7) above, the same effects as those of the information analysis device (1) above can be achieved. Effect of the Invention

[0020] According to the present invention, there is provided a news information analysis device, a news information analysis program, and a news information analysis method that acquire a plurality of pieces of news information from a communication network and provide the news information to users for viewing, and that, for a user who frequently accesses news information, calculates a similarity between the news information viewed by the user and other news information not viewed by the user, and recommends to the user other news information not viewed by the user that has a similarity equal to or less than a predetermined value. This makes it possible for even a user who frequently accesses news information to efficiently view articles in a wide range of genres that the user has not viewed. [Brief description of the drawings]

[0021] [Figure 1] 1 is a schematic diagram illustrating a news information analysis system according to an embodiment of the present invention. [Diagram 2] 1 is a block diagram showing a basic configuration of a news information analysis device according to an embodiment of the present invention. [Diagram 3] 4 is a flowchart showing an operation of the news information analysis device in the embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] A preferred embodiment of the news information analysis device of the present invention will be described below with reference to the drawings. Fig. 1 shows the basic configuration of a news information analysis system 100. Note that in this embodiment, the number of news information items with different news identification information viewed by each user is exemplified as the number of views of news information tallied for each user. Hereinafter, the number of news information items with different news identification information will simply be referred to as the number of news information items, unless otherwise specified. 1, the news information analysis system 100 includes an external system 5, a news information analysis device 1, and a communication network 60. Note that the system may also include a client terminal 2 (an information processing device such as a mobile terminal, a tablet terminal, or a PC). Examples of external systems 5 include web media sites (primary media sites), web media sites (secondary media sites), technology and business information media sites, financial information media sites, web media sites (social media sites), government agency media sites, corporate media sites, research institute media sites, etc. The news information analysis device 1 acquires multiple pieces of news information distributed on the Internet by an external system 5, assigns news identification information to each piece of news information, and provides the news information to users for viewing, while also identifying users who view the news information, tallying up the number of pieces of news information viewed by each user, and for users who frequently access news information, calculating the similarity between the news information viewed by the user and news information not viewed by the user, and recommending to the user news information not viewed by the user and having a similarity below a preset predetermined value. This makes it possible for even users who frequently access news information to efficiently view articles in a wide range of genres. For example, for a user who frequently accesses news information, the client terminal 2 can calculate the similarity between the news information viewed by the user and news information that the user has not viewed, and output news information that the user has not viewed and whose similarity is equal to or less than a preset value to the client terminal 2 to which the user is logged in. This allows even a user who frequently accesses news information to efficiently view articles in a wide range of genres. The above has described an outline of the system configuration in the present embodiment 1. Next, the news information analysis device 1 will be described.

[0023] As shown in FIG. 2, the news information analysis device 1 includes a control unit 11, a storage unit 12, a communication unit 13, a display unit 14, and an input unit 15.

[0024] Control unit 11 is composed of an arithmetic processing device such as a microprocessor, and controls each unit constituting news information analysis device 1. Details of control unit 11 will be described later.

[0025] The storage unit 12 is composed of a semiconductor memory and the like, and stores various information such as control programs called firmware or operating systems, programs for causing the control unit 11 to function as news information acquisition functions, output functions, user information management functions, similarity calculation functions, output control functions, and the like, connection information required for connection to an external system 5 from which news information is acquired via a communication network 60, news information acquired from the external system 5, news identification information for identifying each piece of acquired news information, and a set of news identification information of news information actually accessed by a user for each user identification information, which is linked to the user identification information of the user (hereinafter also referred to as "user-specific viewed news information"), among other news information provided to a user for viewing via the client terminal 2. In this embodiment, a login ID is exemplified as the user identification information. For this reason, in addition to the program, the memory unit 12 is equipped with an external system connection information memory unit 120 which contains connection information necessary for connecting to an external system 5 from which news information is obtained, a news information storage memory unit 121 which stores news information obtained from each external system 5, and a user-specific viewed news information memory unit 122 which stores a set of news identification information of news information actually viewed by a user for each user identification information ("user-specific viewed news information").

[0026] The external system connection information storage unit 120 stores information required for connecting to an external system 5 that acquires news information via the communication network 60 (e.g., a connection URL, authentication information, timing information for acquiring newly created and / or updated news information from the external system 5, etc.). As mentioned above, examples of the external system 5 include mass media sites, technology and business information media sites, financial information media sites, corporate sites, research institute sites, government agency sites, social media sites, etc. News information is not limited to Japanese, and news information in English, Chinese, and Korean may also be acquired. In this way, the news information acquisition unit 110, which will be described later, may refer to the external system connection information storage unit 120, connect to each external system 5 at a predetermined timing (e.g., time information, receiving notifications of information creation and update from the external system 5, etc.), and acquire new and / or updated news information provided by the external system 5.

[0027] The news information storage memory unit 121 is a memory unit that stores new and / or updated news information acquired by the news information acquisition unit 110 from each external system 5 after assigning news identification information that identifies the news information.

[0028] The user-specific viewed news information memory unit 122 is a memory unit that links the news identification information of news information actually viewed by a user among the news information provided to the user for viewing by the news information analysis device 1 to the user identification information of the user, and stores a set of news identification information of news information actually viewed by the user for each user identification information ("user-specific viewed news information"). As described later, the user-specific viewed news information may be a set of news identification information of news information actually viewed by the user during the period from when the user logs in to the news information analysis device 1 until when the user logs out. The user-specific viewed news information may also be a set of news identification information of news information actually viewed by the user during the period from when the user logs in to the news information analysis device 1 until a preset specified time has elapsed. The user-specific viewed news information may also be a set of news identification information of news information actually viewed by the user after logging in to the news information analysis device 1 within a one-day period. The storage unit 12 has been described above.

[0029] The communication unit 13 has a DSP and the like, and realizes wireless communication and wired communication with the external system 5 via the communication network 60 in compliance with standards such as LTE (Long Term Evolution), 4G (4th Generation), and 5G (5th Generation) and Wi-Fi (registered trademark). The communication network 60 is realized by a network such as the Internet or a mobile phone network, or a network that combines these. In addition, a LAN (Local Area Network) may be included as part of the network.

[0030] The display unit 14 is configured with a display device such as a liquid crystal display or an organic electroluminescence panel. The display unit 14 displays an image upon receiving an instruction from the control unit 11. For example, the display unit 14 may display a message to be output to an operator. The input unit 15 is composed of an input device (not shown) such as a physical switch called a numeric keypad or a touch panel provided over the display surface of the display unit 14. For example, an operator may input processing commands and the like to the control unit 10 via the input unit 15.

[0031] Next, the control unit 11 will be described in detail. The control unit 11 is composed of a microprocessor having a CPU, RAM, ROM, I / O, etc. The CPU executes each program read from the ROM or storage unit 12, and during execution, reads information from the RAM, ROM, and storage unit 12, writes information to the RAM and storage unit 12, and transmits and receives signals to and from the communication unit 13. In this way, the processing in this embodiment is realized by the cooperation of the hardware and software (programs).

[0032] The control unit 11 includes, as functional blocks, a news information acquisition unit 110, an output unit 111, a user information management unit 112, a similarity calculation unit 113, and an output control unit 114. Equipped with.

[0033] The news information acquisition unit 110 connects to each external system 5 that distributes news information (title, text, photos, drawings, etc.) via the communication network 60 at a predetermined timing based on a connection URL, authentication information, etc., and acquires newly created news information or updated news information distributed from each external system 5. Here, examples of the predetermined timing include a preset time, a preset time interval, or receipt of a distribution notification from each external system 5 (including receipt of breaking news, special editions, etc. in addition to news information that is distributed periodically). As described above, the news information distributed from the external system 5 is not limited to Japanese, and may be acquired in other languages ​​such as English, Chinese, Korean, etc., as required by the user. In this case, the news information acquisition unit 110 may generate a Japanese translation in addition to the original language. The news information acquisition unit 110 assigns news identification information to each piece of news information acquired from each external system 5, and stores the news information linked by the news identification information in the news information storage memory unit 121. Note that the news information linked by the news identification information may be stored in the news information storage memory unit 121 with the site name of the external system 5 from which the news information was acquired and acquisition timing information (for example, distribution time information (date and time)).

[0034] The output unit 111 provides news information stored in the news information storage memory unit 121 for users to view. Each user has user identification information (e.g., a login ID), and by logging in and accessing the news information analysis device 1, for example, via the client terminal 2, the output unit 111 provides the news information stored in the news information storage memory unit 121 for the user to view (via the client terminal 2). The output unit 111 provides the news information stored in the news information storage memory unit 121 for the user to view via the client terminal 2. Note that the output unit 111 may provide a list of news information assigned the same news identification information to each user as a top news screen (hereinafter also referred to as a "default viewing and provision screen") for each user to view. Furthermore, the output unit 111 may be controlled by the output control unit 114 described later so as to recommend to a user who frequently accesses news information other news information not viewed by the user, the news information viewed by the user having a similarity to the other news information not viewed by the user that is equal to or lower than a predetermined value as calculated by the similarity calculation unit 113 described later. The detailed functions of the similarity calculation unit 113 and the output control unit 114 will be described later.

[0035] The user information management unit 112 links the news identification information of news information actually viewed by the user among the news information provided to the user for viewing by the output unit 111 via the client terminal 2 to the user identification information of the user (e.g., login ID), and stores a set of news identification information of news information actually viewed by the user for each user identification information ("user-specific viewed news information") in the user-specific viewed news information storage unit 122. As described above, when creating the user-specific viewed news information, the user information management unit 112 may set as the user-specific viewed news information a set of news identification information of news information actually viewed by the user during a period from when the user logs in to the news information analysis device 1 until when the user logs out. Also, when creating the user-specific viewed news information, the user information management unit 112 may set as the user-specific viewed news information a set of news identification information of news information actually viewed by the user during a period from when the user logs in to the news information analysis device 1 until a predetermined time has elapsed. Also, the user information may set as the user-specific viewed news information a set of news identification information of news information actually viewed by the user after logging in to the news information analysis device 1 within a period of one day. Note that the predetermined period, the predetermined period range, and the like are not limited to the above examples. Also, an arbitrary period may be set as the predetermined period as necessary. The user information management unit 112 may refer to the user-specific news information viewed storage unit 122 and calculate, for each user identification information, the number of news information viewed by the user within a specified period based on the user-specific news information viewed.

[0036] Before describing the details of the function of the similarity calculation unit 113, overlapping and non-overlapping parts between texts of news information will be described. Generally, news information is characterized by descriptions in concise sentences, and news information about the same event is often described in the same way. For this reason, news information is considered to first have an objective description about the same event, followed by text (e.g., evaluation, etc.) that reflects the writing attitude (e.g., positive, negative, or neutral) of the news information distributor or reporter about the event. In this way, objective descriptions about the same event are considered to be texts with a high degree of agreement (referred to as "overlapping parts") regardless of the external system 5 that distributes the news. In fact, there are many cases where objective descriptions about the same event in news information are quoted. On the other hand, texts that reflect the writing attitude (e.g., positive, negative, or neutral) of the news information distributor or reporter about the event are considered to be texts with a low degree of agreement (referred to as "non-overlapping parts"). In addition, it is considered that the distribution of news information about the same event is concentrated at nearby times (time periods). In this way, by estimating the overlapping and non-overlapping parts between any two pieces of news information delivered at nearby times (time periods), it is possible to determine the similarity between the news pieces of information according to the amount of overlapping parts. For example, if there is a large amount of overlap between the text of news information A and the text of news information B, it can be assumed that news information A and news information B are news about the same event. Conversely, if there is a small amount of overlap between the text of news information A and the text of news information C, it can be assumed that news information A and news information C are news about different events.

[0037] As mentioned above, objective descriptions of the same event are considered to be texts with a high degree of similarity (referred to as "overlapping portion") regardless of the external system 5 that distributes the news. For this reason, the news information (text) acquired from each external system 5 whose news information distribution date and time falls within a predetermined period range is treated as the target (hereinafter referred to as the "similarity determination target set"), and the overlapping portion between any two pieces of text information of different news information (for simplicity, referred to as "news information A" and "news information B") included in the similarity determination target set can be calculated based on the degree of similarity between the text documents. Here, as a method for determining the degree of similarity between news information A (text A) and news information B (text B), a method known to those skilled in the art may be applied, for example, a method of setting a similarity threshold for indicators such as the number of matching characters, edit distance, and the matching rate of words divided into n-grams, and making a judgment. For example, the degree of similarity may be calculated based on the degree of match of character strings obtained by n-gram-dividing all of the texts A and B in sentence or paragraph order from the start of the main text, and if the degree of similarity is below or equal to the similarity threshold, it may be determined that the news is about different events. Also, all of the texts A and B are morphologically analyzed in sentence order or paragraph order from the start of the main text, and the Jaccard coefficient or the like is used to determine the degree of agreement between words (e.g., nouns and verbs) that make up the content of the sentence. If the degree of agreement is below or equal to a similarity threshold, it may be determined that the news is about a different event. Also, text information may be compared as a method of determining the degree of agreement. In this way, the overlapping portion between each piece of news information and other news information can be estimated based on the degree of agreement.

[0038] The text portions that match between two pieces of news information belonging to a similarity determination target set can be classified as overlapping portions of the two pieces of news information, and the portions of the news information excluding the overlapping portions can be classified as non-overlapping portions. The similarity between the pieces of news information can then be determined according to the amount of overlapping portions. Here, the amount of overlapping portions may be, for example, the number of characters in the text of the overlapping portions, the ratio of the overlapping portions to the non-overlapping portions, or the proportion of the overlapping portions in the total text of the news information. In this way, the similarity can be calculated for each pair of any two pieces of news information included in the similarity determination target set. The above describes overlapping and non-overlapping parts between texts of news information. Next, a description will be given of the similarity calculation unit 113 that calculates the similarity between news information viewed by a user who frequently accesses news information and other news information not viewed by the user, for the user who has viewed the news information and has a number of viewed news information items equal to or greater than a preset threshold.

[0039] The similarity calculation unit 113 identifies users who frequently access news information based on the news information viewed by each user. When there are multiple users who frequently access news information, the similarity calculation unit 113 calculates the similarity between the news information viewed by each user who frequently accesses news information and the news information not viewed by that user. The similarity calculation unit 113 identifies users who frequently access news information one by one, and calculates the similarity between the news information viewed by that user and the news information not viewed by that user. For this reason, the similarity calculation unit 113 identifies a set of news information viewed by a user based on the viewed news information for each user (hereinafter referred to as "viewed news information"), and extracts the complementary set of news information that the user did not view (hereinafter referred to as "unviewed news information"). The similarity calculation unit 113 calculates the similarity between each piece of news information included in the viewed news information viewed by the user and the news information included in the unviewed news information, and calculates a subset of unviewed information whose similarity is below a predetermined value (referred to as the "first similarity threshold") or does not exceed the predetermined value. In this way, the similarity calculation unit 113 can calculate, for each news information included in the viewed news information viewed by the user, a subset of unviewed information whose similarity is less than the first similarity threshold or does not exceed the first similarity threshold. The similarity calculation unit 113 can calculate unviewed news information that has not been viewed by the user and whose similarity to all of the news information viewed by the user is below the first similarity threshold or does not exceed the first similarity threshold (hereinafter referred to as ``unviewed news information with low similarity''), for example, by calculating the intersection of a subset of unviewed information whose calculated similarity for each news information included in the viewed news information viewed by the user is below the first similarity threshold or does not exceed the first similarity threshold. The similarity calculation unit 113 may further register a news information group ID that constitutes the same event news information for the news information registered and stored in the news information storage memory unit 121, for example, so that news information that is associated with the same event news information can be identified. The similarity calculation unit 113 can specify whether the news information is of the same event or different events when searching and querying the news information storage memory unit 121. Note that for any pair of two pieces of news information, if the similarity calculated by the similarity calculation unit 113 is below a predetermined similarity threshold value set in advance or is equal to or less than the predetermined similarity threshold value, the two pieces of news information may be determined to be different event news.

[0040] Furthermore, the similarity calculation unit 113 may assign a priority to news information that has not been viewed by a "user who frequently accesses news information" and that is "unviewed news information with low similarity" by giving priority to new news information acquired by the news information acquisition unit 110 at the timing (e.g., distribution time information (date and time) etc.). In this way, the output control unit 114 described later can calculate the news information with high priority in priority when recommending to a "user who frequently accesses news information" news information that has not been viewed by the user and that is "unviewed news information with low similarity". In this way, the output control unit 114 described later can enable a "user who frequently accesses news information" to efficiently view fresh articles in a wide range of genres.

[0041] Furthermore, the similarity calculation unit 113 may preferentially calculate, as recommended high priority news information, "unviewed news information with low similarity" that has not been viewed by a "user who frequently accesses news information" and for which the total number of times the "unviewed news information with low similarity" has been viewed by other users within a specified period is equal to or exceeds a preset value. In this way, the output control unit 114, which will be described later, can enable a "user who frequently accesses news information" to view "unviewed news information with low similarity" that has been viewed frequently by other users.

[0042] Furthermore, the similarity calculation unit 113 may further calculate similarity between low-similarity news information that was not viewed by "users who frequently access news information" so as to preferentially calculate the news information with the largest amount of information (e.g., amount of characters) in the overlapping part among the high-similarity news information whose similarity is equal to or exceeds a preset predetermined value. Note that the overlapping part of news information A is considered to depend on news information B to be compared, but as described above, by classifying news information into same-event news information groups, the overlapping part of news information A may be the largest of the overlapping parts calculated by comparison with each news information belonging to the same event information group. In this way, the output control unit 114, which will be described later, can enable "users who frequently access news information" to view "unviewed news information with low similarity" that has the largest amount of overlapping information, for example.

[0043] In addition, the similarity calculation unit 113 may be configured to calculate "unviewed news information with low similarity" that has not been viewed by a "user who frequently accesses news information" by giving priority to the unviewed news information with lower similarity as high priority news information to be recommended.

[0044] Furthermore, when "unviewed news information with low similarity" that has not been viewed by a "user who frequently accesses news information" is the same or similar news information viewed by another "user who frequently accesses news information," the similarity calculation unit 113 may preferentially calculate the "unviewed news information with low similarity" as high priority news information to be recommended to the "user who frequently accesses news information." In this way, the output control unit 114, which will be described later, can make a wide range of "unviewed news information with low similarity" viewable by "users who frequently access news information."

[0045] The output control unit 114 controls the output unit 111 to recommend to each "user who frequently accesses news information" news information that has not been viewed by the user, as calculated by the similarity calculation unit 113, and that has a low similarity compared to the news information viewed by the user ("unviewed news information with low similarity"), for example, news information whose similarity to the viewed news information is below a predetermined value set in advance. This allows even a user who frequently accesses news information to view news information that is less similar to the news information that he or she has viewed, enabling efficient viewing of articles in a wide range of genres. Furthermore, as mentioned above, by making it possible to view "unviewed news information with low similarity" depending on the calculation method used by the similarity calculation unit 113, it is possible, as mentioned above, even for a user who frequently accesses news information, to efficiently view articles in a wide range of genres that the user has not viewed. The configuration of each functional unit of the news information analysis device 1 exemplified as the present embodiment has been described above.

[0046] Next, the operation of this embodiment will be described with reference to the flowchart shown in Fig. 3. Fig. 3 is a flowchart showing the operation of the news information analysis device 1. The process shown in Fig. 3 illustrates an example of a processing form in which the output unit 111 is controlled to recommend to a "user who frequently accesses news information" within a predetermined time period news information that has not been viewed by the user and has a low similarity compared to the news information that the user has viewed.

[0047] Referring to Figure 3, in step S10, the news information analysis device 1 (news information acquisition unit 110) connects to each external system 5 that distributes news information via the communication network 60 at a predetermined timing based on a connection URL, authentication information, etc., and acquires newly created news information or updated news information distributed from each external system 5, and assigns news identification information to each piece of news information and stores it in the news information storage memory unit 121.

[0048] In step S11, the news information analysis device 1 (output unit 111) provides the news information stored in the news information storage memory unit 121 for viewing by the user.

[0049] In step S12, the news information analysis device 1 (user information management unit 112) links the news identification information of the news information actually viewed by the user among the news information provided to the user for viewing by the output unit 111 to the user identification information of the user, and stores a set of news identification information of the news information actually viewed by the user for each user identification information ("user-specific viewed news information") in the user-specific viewed news information storage unit 122.

[0050] In step S13, the news information analysis device 1 (user information management unit 112) refers to the user-specific viewed news information storage unit 122 and calculates, for each user identification information, the number of news information items actually viewed by the user.

[0051] In step S14, the news information analysis device 1 (similarity calculation unit 113) identifies, for each user identification information, user identification information for which the number of news information items actually viewed by the user within a specified period of time is equal to or exceeds a preset threshold value ("users who frequently access news information").

[0052] In step S15, the news information analysis device 1 (similarity calculation unit 113) calculates, for each frequently accessed user, a similarity between news information viewed by the user and news information not viewed by the user.

[0053] In step S16, the news information analysis device 1 (similarity calculation unit 113) calculates news information that has not been viewed by the user and whose similarity to the news information viewed by the user is below a predetermined value set in advance ("unviewed news information with low similarity").

[0054] In step S17, the news information analysis device 1 (output control unit 114) controls the output unit 111 to recommend, for each user with high access frequency, the "unviewed news information with low similarity" calculated in step S16.

[0055] The above has explained the operation of the news information analysis device 1, which acquires multiple news information distributed on the Internet by the external system 5, assigns news identification information to each piece of news information, and then provides the news information to users for viewing, as well as identifying the users viewing the news information, tallying up the number of pieces of news information viewed by each user, and controlling the output to recommend to ``users who frequently access news information'' for whom the number of pieces of news information viewed is equal to or exceeds a predetermined threshold, news information that the user did not view but that has a low similarity compared to the news information viewed by the user. As described above, the news information analysis device 1 can enable even a user who frequently accesses news information to efficiently browse articles in a wide range of genres.

[0056] Each of the devices can be realized by hardware, software, or a combination of these. In addition, the navigation method performed by the cooperation of each of the devices included in the navigation system can also be realized by hardware, software, or a combination of these. Here, "realized by software" means that it is realized by a computer reading and executing a program.

[0057] The program can be stored and provided to the computer using various types of non-transitory computer readable media. The non-transitory computer readable media includes various types of tangible storage media. Examples of the non-transitory computer readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROM (Read Only Memory), CD-R, CD-R / W, and semiconductor memory (e.g., mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (random access memory)). The program may also be provided to the computer by various types of transitory computer readable media. Examples of the transitory computer readable media include electric signals, optical signals, and electromagnetic waves. The transitory computer readable media can provide the program to the computer via a wired communication path such as an electric wire and an optical fiber, or via a wireless communication path.

[0058] The above-described embodiment is a preferred embodiment of the present invention, but the scope of the present invention is not limited to the above-described embodiment, and the present invention can be implemented in various modified forms without departing from the gist of the present invention.

[0059] <Variation 1> In the above embodiment, the user information management unit 112 identifies the user by, for example, a login ID, but this is not limiting. For example, the user information management unit 112 may use cookie information to identify the user.

[0060] <Variation 2> In the above-described embodiment, the number of news information items with different news identification information viewed by each user is exemplified as the number of times the news information is viewed by each user, the number of times being determined as the number of times the news information items with different news identification information are viewed by the user within a specified period. In contrast, instead of the number of times the news information items with different news identification information are viewed as the number of times the news information is accessed, the number of times the news information is accessed may be the number of times the news information is accessed, the number of times being determined as the number of times the news information is accessed, the number of times being determined as the number of times the news information is accessed, the number of times being determined as the number of times. Specifically, when the same news information is accessed multiple times (n times; n>1), the number of news information items with different news identification information is calculated as 1, but the number of times the news information is accessed is calculated as n.

[0061] <Modification 3> In the above-described embodiment, the number of news information items with different news identification information viewed by each user was exemplified as the number of times the news information was viewed by each user, as the number of times the news information was viewed by each user, as tabulated for each user. In contrast, for example, the number of times each user logged in to a news information providing site provided by the news information analysis device 1 within a predetermined period of time may be the number of times the news information was viewed by each user, as tabulated for each user. In this case, for example, if a user logs in but does not access any news information within a preset time period, the user may be automatically logged off and the number of accesses may be set to zero. For example, users who access the site a specified number of times or more each day may be designated as rank A, those who access the site every day as rank B, those who access the site four or more times a week as rank C, those who access the site two to three times a week as rank D, and those who access the site once a week or less as rank E. In this case, for example, users with ranks A, B, and C may be determined as users with high access frequency.

[0062] <Modification 4> In the above embodiment, the news information analysis device 1 is described as being realized by one server device or the like, but the functions of the news information analysis device 1 may be appropriately distributed across multiple server devices to form a distributed processing system. Furthermore, each function of the news information analysis device 1 may be realized by using a virtual server function or the like on a cloud. [Explanation of symbols]

[0063] 100 News Information Analysis System 1. News information analysis device 11 Control section 110 News Information Acquisition Department 111 Output section 112 User Information Management Department 113 Similarity calculation unit 114 Output control section 12 Storage section 120 External system connection information storage unit 121 News information storage unit 122 User-specific news browsing information storage unit 13. Communications Department 14 Display section 15 Input section 2. Information processing terminal (client terminal) 5 External Systems 60 Communication Network

Claims

1. a news information acquiring unit that acquires a plurality of pieces of news information including text information from a communication network, assigns news identification information to each piece of news information, and stores the news information; an output unit that provides the news information stored by the news information acquisition unit to a user for viewing; a user information management unit that identifies the users who view the news information provided for viewing by the output unit and tally up the number of times the news information has been viewed for each user; a similarity calculation unit that calculates a similarity between the news information, which is assigned with the news identification information and which the user has viewed, and other news information that the user has not viewed, for the user whose number of views of the news information is equal to or greater than a preset threshold value; an output control unit that controls the output unit so as to recommend to the user other news information that has not been viewed by the user and whose similarity is equal to or less than a predetermined value that is set in advance; A news information analysis device comprising:

2. The news information analysis device of claim 1, characterized in that the number of times the news information is viewed is either the number of news information with different news identification information accessed by each user, the number of times the news information accessed by each user is accessed, or the number of times a news information providing site that provides the news information for viewing by each user is accessed.

3. 3. The news information analysis device according to claim 1, wherein the similarity calculation section calculates the similarity based on a degree of similarity between the text information items included in the news information.

4. The news information analysis device according to claim 1 or 2, characterized in that the output control unit prioritizes the most recent news information acquired by the news information acquisition unit with respect to the news information recommended to the user.

5. The news information analysis device of claim 1 or 2, characterized in that the output control unit preferentially recommends to the user news information having news identification information whose number of views by users other than the user is equal to or greater than a preset value, for other news information whose similarity is equal to or less than a preset value and whose number of views by users other than the user is equal to or greater than a preset threshold value and whose number of views is equal to or greater than a preset threshold value and whose number of views is equal to or less than a preset value.

6. A computer device connected to a communications network, a news information acquiring unit that acquires a plurality of pieces of news information including text information from a communication network, assigns news identification information to each piece of news information, and stores the acquired news information; an output unit for providing the news information stored by the news information acquisition unit to a user for viewing; a user information management unit that identifies the users who view the news information provided for viewing by the output unit and tally up the number of times the news information has been viewed for each user; a similarity calculation unit that calculates a similarity between the news information accessed by the user, the news information being assigned with the news identification information, and other news information not accessed by the user, for the user whose number of accesses to the news information is equal to or exceeds a preset threshold value; an output control unit that controls the output unit so as to recommend to the user other news information that has not been viewed by the user and whose similarity is equal to or less than a predetermined value that is set in advance; A news information analysis program characterized by functioning as:

7. 1. A computer-implemented method for analyzing news information, comprising: a news information acquisition step of acquiring a plurality of pieces of news information including text information from a communication network, assigning news identification information to each piece of news information, and storing the acquired news information; an output step of providing the news information stored in the news information acquisition step to a user for viewing; a user information management step of identifying the users who view the news information provided for viewing in the output step and tallying up the number of times the news information has been viewed for each user; a similarity calculation step of calculating a similarity between the news information attached with the news identification information and viewed by the user, the news information being viewed by the user having a number of views of the viewed news information equal to or greater than a preset threshold value, and other news information not viewed by the user; an output control step of controlling the output step so as to recommend to the user other news information that has not been viewed by the user and whose similarity is equal to or less than a predetermined value set in advance; A news information analysis method comprising:

Images