Content providing system and the like

Abstract

To provide a content providing system and the like which provide a user with content considering a scene or a state during moving.SOLUTION: A content providing system 10 comprises: a content storage unit 15 in which content lists 15a with which keywords 15a are associated are stored; a static characteristic storage unit 12 in which static characteristic information 12a including geographic information of divided areas 9 into which a map is divided into multiple areas is stored; a position information acquisition unit 1 which acquires position information 1a of a terminal 17; a determination unit 29 which determines area characteristic information 14a on the basis of static characteristic information of a divided area corresponding to the position information and dynamic characteristic information 13a including a hot word 21a based on posted SNS information about the divided area corresponding to the position information; a selection unit 2 which selects a corresponding content list on the basis of the area characteristic information and a keyword; and a providing unit 4 which provides the selected content list to the terminal.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a content providing system that provides content to a user based on a current location. [Background technology]

[0002] For example, Patent Document 1 discloses an invention of a music selection support system that enables selection of music suited to the current location of a vehicle based on current location information acquired by an on-board navigation device.

[0003] Furthermore, Patent Document 2 discloses an invention that extracts important and characteristic phrases from among phrases (names or attributes of locations contained in each area) associated with multiple areas that include a travel route from a starting point to an end point, and selects content related to the phrases.

[0004] Furthermore, conventionally, the music preferences of a user are learned and similar music is reproduced in the vehicle. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2006-133618 A [Patent Document 2] Patent No. 4929225 Summary of the Invention [Problem to be solved by the invention]

[0006] In the past, content was selected mechanically according to one's preferences based on location information / travel route. However, even if one travels the same route, emotions felt by people may differ due to, for example, seasonal changes, and the vehicle's condition, such as the vehicle's speed, and the surrounding environmental conditions, such as weather, temperature, and time (morning / afternoon / evening), may also differ from day to day. On the other hand, since the atmosphere of urban and rural areas is very different, it is preferable to respect regional characteristics such as the area's geography and features.

[0007] Therefore, an object of the present invention is to provide a content providing system that provides a user with content that takes into account the scenery and conditions during travel. [Means for solving the problem]

[0008] (1) One embodiment of the content provision system of the present invention is characterized by comprising a content memory unit that stores a content list associated with keywords, a static characteristic memory unit that stores static characteristic information including geographic information of a divided area obtained by dividing a map into a plurality of areas, a location information acquisition unit that acquires location information of a terminal, a determination unit that determines area characteristic information based on the static characteristic information of the divided area corresponding to the location information and dynamic characteristic information including hot words based on posted SNS information related to the corresponding divided area, a selection unit that selects the corresponding content list based on the area characteristic information and the keywords, and a provision unit that provides the selected content list to the terminal.

[0009] "Location information" is information that indicates the current location of a moving object. "Location information" is, for example, information expressed as latitude and longitude. "Location information" may also be information expressed with altitude added, or may be information expressed as another index value that can be converted to "latitude, longitude, altitude."

[0010] "Keywords" are, for example, words that indicate the image or atmosphere of the content list, words that indicate characteristics, words that are included in the content list, words that are related to the content list, or co-occurrences of these words. "Keywords" may also be information that is expressed as a value that can be converted into these examples of words.

[0011] - A "content list" is information that bundles together one or more pieces of content consisting of audio, images, videos, or a combination of these. A "content list" may also be information that includes values ​​that can be converted into these pieces of content.

[0012] "Static characteristic information" refers to words / phrases or related words that indicate the geography or features associated with a particular zoning area. "Static characteristic information" may also refer to words / phrases or related words that indicate the culture, history, people, or brand image associated with a zoning area. "Static characteristic information" may also refer to information that includes values ​​that can be converted to these words / phrases. For example, geography is a term that refers to the distribution of land, sea, mountains, rivers, climate, living things, population, cities, industry, transportation, and other conditions of the land. For example, a feature is a word / phrase that describes an object on the ground, whether natural or man-made, such as a building, tree, rock, etc.

[0013] "Dynamic characteristic information" refers to information on social media posts, web pages, magazines, television and radio programs, search words, or other related information related to a specific area. "Dynamic characteristic information" may also be information that includes values ​​and data that can be converted into these terms.

[0014] "Area characteristic information" is information obtained based on static characteristic information and dynamic characteristic information. For example, the "area characteristic information" may be information obtained by simply concatenating the static characteristic information and the dynamic characteristic information, or by performing natural language processing on the "area characteristic information." The natural language processing may be, for example, information obtained by vectorizing the static characteristic information and the dynamic characteristic information and calculating the average value.

[0015] (2) In such a content providing system, it is preferable that the selection unit is further based on mobile object information including the status of the mobile object holding the terminal.

[0016] "Mobile object information" is information about the state of a mobile object. "Mobile object information" is information that indicates the state of a mobile object, such as images from a drive recorder, speed, acceleration, number and frequency of braking, acceleration, wiper operation, etc. "Mobile object information" may also be information that indicates the state of the surroundings of a mobile object, such as temperature, humidity, weather, traffic congestion, etc. Furthermore, "mobile object information" may be information related to these pieces of information, or may be information that includes values ​​that can be converted into these pieces of information.

[0017] (3) It is also preferable that the information processing device further comprises a generating unit that generates display elements related to the content list for display on a map screen displayed on the terminal.

[0018] "Display elements" are elements that display information related to a content list. "Display elements" may be, for example, text / illustrations of the aforementioned keywords, the selected origin of the selected content list, i.e., static characteristic information, dynamic characteristic information, or area characteristic information. Furthermore, "display elements" may be information related to these, or information that includes values ​​that can be converted into these pieces of information.

[0019] (4) A server according to another aspect of the present invention is characterized in that area characteristic information is set based on static characteristic information including geographic information of a divided area obtained by dividing a map into a plurality of areas, and dynamic characteristic information including hot words based on SNS information posted about the divided area, and the server is equipped with a location information acquisition unit that acquires location information of a terminal, and a selection unit that selects a corresponding content list to be provided to the terminal based on keywords associated with the area characteristic information and a content list corresponding to the location information.

[0020] (5) Another aspect of the program of the present invention is a program executed by a server, in which area characteristic information is set based on static characteristic information including geographic information of a divided area obtained by dividing a map into a plurality of areas, and dynamic characteristic information including hot words based on SNS information posted about the divided area, and which is characterized in that it acquires location information of a terminal, and selects the corresponding content list to be provided to the terminal based on keywords associated with the area characteristic information and content list corresponding to the location information.

[0021] (6) Another aspect of the content selection method of the present invention is a content selection method executed by a computer, which is characterized in that static characteristic information including geographic information of a divided area obtained by dividing a map into a plurality of areas is set in advance, area characteristic information is determined based on the static characteristic information of the divided area corresponding to a user's location information and dynamic characteristic information including hot words based on posted SNS information regarding the divided area corresponding to the location information, and a corresponding content list to be provided to the terminal is selected based on keywords associated with the area characteristic information and the content list.

[0022] (7) A server according to another aspect of the present invention is characterized in that it is partitioned into a plurality of areas, a content list related to the partitioned areas is set, and the server is provided with a generation unit that generates display elements related to the content list for display in the partitioned areas of a map screen displayed on a terminal.

[0023] (8) Such a server is preferably arranged so that the display elements are aligned along roads displayed on the map screen. Effect of the Invention

[0024] The content providing system can provide different content according to the situation, and can prevent the user from getting bored with the content. [Brief description of the drawings]

[0025] [Figure 1] FIG. 1 is a schematic diagram illustrating an embodiment of a content providing system. [Diagram 2] FIG. 1 is a functional block diagram illustrating an embodiment of a content providing system. [Diagram 3] FIG. 4 is a schematic diagram illustrating an example of a data structure of a static characteristic information database. [Figure 4] FIG. 4 is a schematic diagram illustrating an example of a data structure of a dynamic characteristic information database. [Diagram 5] FIG. 4 is a schematic diagram illustrating an example of a data structure of a content list. [Figure 6] FIG. 2 is a schematic diagram illustrating an example of a hardware configuration of a server. [Figure 7] FIG. 2 is a schematic diagram illustrating an example of a hardware configuration of a terminal. [Figure 8] 11 is a flow chart illustrating one embodiment of a server process flow. [Figure 9] FIG. 2 is a schematic diagram illustrating an example of a data structure of an area characteristic keyword database. [Figure 10] FIG. 4 is a schematic diagram illustrating an example of a data structure of a display element. [Figure 11] FIG. 4 is a schematic diagram showing an example of a map screen displayed on the terminal. [Figure 12] FIG. 11 is a schematic diagram showing another example of a map screen displayed on the terminal. [Figure 13] FIG. 2 is a schematic diagram showing an example of a map screen displayed on a terminal or a mobile object. [Figure 14] This is an enlarged map screen of point A in FIG. 13. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] [1. Overview] (Introduction) The content provision system of this embodiment, roughly speaking, matches static characteristic information that embodies the characteristics of an area, such as the geography and features that have been established in a specific region (area) over a relatively long span of time and the geography and features of the area that are likely to be established in the future, dynamic characteristic information that is, so to speak, "in season," such as movements / trends that are a topic of conversation for a limited period of time related to the area, with a content list of music, for example by language analysis, and provides a content list that takes into account the regional characteristics and the current trends to people / vehicles / etc. located in the area. In other words, it provides content selected by "Now! If it's in that area!" Therefore, even if it is the same area, it is possible to provide a different content list every day that takes into account the locality (location) and seasonal (time) elements, so that users do not get bored.

[0027] (Content Providing System 10) First, a content providing system (hereinafter, simply referred to as the system) will be described with reference to Fig. 1. Fig. 1 shows a system 10. The system 10 includes, for example, a server 11 that selects a content list to be provided, a static characteristic storage unit 12 (external server) that stores static characteristic information 12a indicating characteristics of a partitioned area 9 (see Fig. 2) that partitions a predetermined area of ​​a map, a dynamic characteristic storage unit 13 (external server) that stores dynamic characteristic information 13a (see Fig. 2) indicating seasonal characteristics within the partitioned area 9, a determination server 14 (external server) that determines and stores area characteristic information 14a (see Fig. 2), a content storage unit 15 (external server), an external device 16 (external server) that receives information such as weather from an external organization, and a terminal 17 / mobile body 18 that receives the content list, each of which is communicably connected via a communication network 19.

[0028] [2.Each configuration] (Server 11) Fig. 2 shows an embodiment of a functional block diagram of the system 10. The server 11 shown in Fig. 2 mainly comprises, for example, a selection unit 2 that selects a content list 15a to be provided. This embodiment also includes an extraction unit 3 that extracts a content list 15a, and a provision unit 4 that provides the selected content list 15a to a terminal 17. Furthermore, it includes a generation unit 5 that generates display elements 6 related to the content list 15a to be displayed on a map screen displayed on the terminal 17. It also includes a display control unit 40 that controls the display position on the map screen 8 (see FIG. 13) of the display elements 6 generated by the generation unit 5.

[0029] (Selection 2) The selection unit 2 selects a content list 15a to be provided to the terminal 17 based on area characteristic information 14a and content keywords 15b in the section area 9 corresponding to the location information 1a transmitted from the terminal 17.

[0030] (Extraction part 3) The extraction unit 3 extracts the selected content list 15 a and stores it in the server 11 .

[0031] (Provider part 4) The providing unit 4 provides the selected content list 15 a to the terminal 17 .

[0032] (Generation part 5) The generation unit 5 generates display elements 6 to be displayed on a map screen 8 displayed on a terminal 17 .

[0033] (Display control unit 40) The display control unit 40 determines the display position of the display element 6 on the map screen 8.

[0034] (Plot Area 9) The divided areas 9 divide the map into areas of a predetermined shape and size. Any area may be divided so as to enclose a predetermined area. In this embodiment, the map is divided into a mesh shape consisting of a plurality of squares. The shape of each mesh may be rectangular or triangular, or may be circular or elliptical. Furthermore, the shapes of the divided areas 9 do not all have to be the same shape. The size of the divided area 9 may be different depending on the location on the map. For example, it may be large in areas with the same scenery, sparsely populated areas, areas with similar geography and features, etc. Conversely, it may be small in areas with large changes in scenery, tourist spots, densely populated areas, areas with a concentration of features such as bridges, buildings, and landmark buildings.

[0035] (Static characteristic storage unit 12) In this embodiment, the static characteristic storage unit 12 is an external data server, and stores static characteristic information 12a for each section area 9. The static characteristic information 12a is organized in a static characteristic database (hereinafter, referred to as a static characteristic DB) 20.

[0036] (Static characteristics DB20) 3 is a schematic diagram showing an example of the data structure of the static characteristic DB 20. The static characteristic DB 20 shown in the figure stores static characteristic information 12a for each divided area 9. The static characteristic information 12a consists of an ID 9a corresponding to the divided area, area characteristic keywords 20a including geography and features, and emotion characteristic keywords 20b. The emotion characteristic keywords 20b are clustered by the emotion characteristic keywords 20b.

[0037] In this embodiment, for example, a large number of songs are classified into the smallest units by morphological analysis, and words that indicate emotional characteristics are extracted. The extracted words are vectorized, clustered, and organized in an emotional characteristics database (hereinafter referred to as emotional characteristics DB) (not shown). For example, emotional characteristics are classified by classifying words such as "cool" and "good" into a group called "cool," and words such as "weak" and "lonely" into a group called "sad."

[0038] (Morphological analysis) In this embodiment, a morphological analysis engine such as MeCab is used. Morphological analysis can classify a sentence into the smallest unit that has meaning in natural language. In morphological analysis, part-of-speech information of classified morphemes can be obtained.

[0039] (Vectorization) In this embodiment, as vectorization of a phrase or a sentence, for example, data obtained by converting morphologically analyzed words into numerical expressions is used as input data to obtain vectors using an algorithm such as word2vec. Vectorization can be calculated using algorithms such as GloVe, WordNet, or fastText in addition to word2vec.

[0040] (Clustering) In this embodiment, the k-means method is used as the clustering method, for example, although other known methods may also be used.

[0041] (Area characteristic keyword 20a) In this embodiment, the area characteristic keywords 20a are extracted from, for example, maps, websites, bulletin boards, digitized official documents and books, and are organized in a geography (land feature) information database (not shown). Note that the area characteristic keywords 20a may be, for example, one or more characteristic words extracted from digitized books such as geography (land features).

[0042] (Feature words) The feature words can be extracted, for example, by the TF-IDF method. The TF-IDF method is a method for extracting words based on the hypothesis that frequently occurring words are important words. Other known methods may also be used.

[0043] 3, in a section area having a waterside and a castle, water, castle, and culture are extracted as area characteristic keywords 20a. Note that the words and sentences on which the extraction is based may be obtained from the Internet by crawling. In this embodiment, the geographic (feature) information DB updates data once a year. However, the data may be updated at a predetermined interval, for example, monthly or every two years or more. Furthermore, the number of area characteristic keywords 20a in this embodiment is three, but it may be less than three or more than three.

[0044] (Emotional trait keyword 20b) The emotion characteristic keywords 20b are obtained from an emotion characteristic database (not shown) that stores keywords indicating emotional characteristics for music. In Fig. 3, the emotion characteristic keywords 20b of quiet, tranquil, and solemn are assigned to the area characteristic keywords 20a of water, castle, and culture, respectively.

[0045] In the present embodiment, one area characteristic keyword 20a corresponds to one emotion characteristic keyword 20b, but multiple emotion characteristic keywords 20b may be associated with each area characteristic keyword 20a. In this case, each emotion characteristic keyword 20b may be weighted.

[0046] (Dynamic characteristic storage unit 13) In this embodiment, the dynamic characteristic storage unit 13 is an external data server. The dynamic characteristic storage unit 13 stores dynamic characteristic information 13a for each partitioned area 9. The dynamic characteristic information 13a is organized in a dynamic characteristic database (hereinafter, referred to as dynamic characteristic DB) 21.

[0047] In this embodiment, a crawler server (not shown) collects information on posted posts from a server on which SNS (social networking service) information is posted. The crawler server extracts spot information including location-related words and characteristic words from the collected posts. In this embodiment, posts from the past 24 hours are collected and spot information is extracted. A frequently occurring characteristic word in the extracted spot information is set as a hot word. Note that hot words may be set for a relatively short period of time, such as monthly, weekly, daily, hourly, or minutely. Furthermore, frequently occurring words other than characteristic words may be set as hot words. There may be one or more hot words.

[0048] Examples of SNS include Twitter (registered trademark), FACEBOOK (registered trademark), INSTAGRAM (registered trademark), YouTube (registered trademark), and LINE (registered trademark). SNS may be broadly interpreted to include blogs or electronic bulletin boards. For example, messages posted by users are published on SNS. For example, on SNS, posters may post their impressions and feelings about various spots such as restaurants and tourist attractions. In addition, on SNS, messages related to the occurrence of incidents and accidents, and traffic congestion may be posted.

[0049] (How to extract words) In this embodiment, for example, machine learning is used to extract words from a sentence. Machine learning is used to infer which keyword type the extracted words belong to. The trained model can infer and classify even an unseen word as to which keyword type it belongs to. The trained model has a function of simultaneously indicating the degree of similarity as a degree of confidence in the analogy of the words, and can also be trained to not adopt a word with a score that is too low.

[0050] (Method of generating trained models for extraction) A method of generating a trained model for extraction is described. For example, a plurality of pieces of data corresponding to the keyword types to be classified, such as "words related to landscapes," "words related to history," and "words related to people," or data that does not correspond to any of them, are prepared in advance as training data in a format paired with keyword type label information. In this case, language data (character data) is handled, and this is vectorized. When a vector of similar training data is input to the trained model, the trained model calculates and outputs the confidence that the input vector matches the classification type. The resulting vector output is a numerical value (confidence) indicating the degree of similarity of the classification type. On the other hand, when a plurality of dissimilar training data are input, a vector is generated that indicates that the corresponding classification does not apply to the training data. The trained model is modified and optimized based on the output result so that the confidence in the label information paired with all of the input data is increased.

[0051] (Other extraction methods) Regarding the extraction method, feature quantities or co-occurring words may be extracted. Furthermore, from the extracted words, words of a predetermined category, such as words related to scenery, may be further extracted.

[0052] (Co-occurring words) Co-occurring words are words that are highly related to the main word and are frequently used together. They can also be said to be words that often appear around the main word, and there are more than one of them. For example, a feature word can be used as the main word.

[0053] (Dynamic characteristics DB21) 4 is a schematic diagram showing an example of the data structure of the dynamic characteristic DB 21. The dynamic characteristic DB 21 shown in the figure stores dynamic characteristic information 13a for each divided area 9. The dynamic characteristic information 13a consists of an ID 9a corresponding to the divided area and a hot word 21a corresponding to the divided area. The figure shows a case where there are many posts on SNS in a specific area about a festival and the opening of a delicious restaurant. In this case, lively, ephemeral, delicious, fun, festival, and food are extracted as hot words 21a. The texts that are the source of the extraction may be obtained from the Internet by crawling. The number of hot words 21a may be less than three or more than three.

[0054] In this embodiment, the divided areas of the static characteristic information 12a and the dynamic characteristic information 13a have the same size and shape. However, the sizes / shapes may be different. In this case, an appropriate divided area is selected based on the position information 1a.

[0055] (Decision Server 14) The determination server 14 includes a determination unit 29 that determines area characteristic information 14a based on the static characteristic information 12a and the dynamic characteristic information 13a, and an area characteristic database (hereinafter, referred to as area characteristic DB) 22 in which the determined area characteristic information 14a is organized. In this embodiment, the determination unit 29 integrates the static characteristic information 12a and the dynamic characteristic information 13a. The area characteristic DB 22 will be described later.

[0056] (Content storage unit 15) The content storage unit 15 is an external data server in this embodiment, and stores a content list 15a. Content list 15a consists of one or more songs, videos, etc. Keywords 15b are set in content list 15a. Keywords 15b characterize content list 15a and consist of one or more keywords. Keywords may include phrases, segments, or sentences.

[0057] (Content List Database 23) 5 is a schematic diagram showing an example of the data structure of a content list database (hereinafter referred to as a content list DB). In the content list DB 23 shown in the figure, content lists 15a and keywords 15b corresponding to the content lists 15a are set. In this embodiment, for example, music genres and emotional characteristics classified in the above-mentioned emotional characteristic DB are described as keywords 15b.

[0058] (External device 16) The external device 16 (see FIG. 1) is an external server that provides the server 11 with external information 16a such as weather and traffic congestion information.

[0059] (Terminal 17) The terminal 17 includes a location information acquisition unit 1. In this embodiment, the terminal 17 is, for example, a personal computer or a smartphone. The terminal 17 may be a mobile phone or a tablet terminal. The terminal 17 may also be a drive recorder or a navigation device. The terminal 17 may also operate in cooperation with a drive recorder or a navigation device.

[0060] (Location information acquisition unit 1) The location information acquisition unit 1 has a function of acquiring location information of the terminal 17. For example, the location information acquisition unit 1 acquires location information using a GPS (Global Positioning System) sensor. Note that a positioning method other than the GPS sensor, such as WiFi positioning or beacon positioning, may also be used.

[0061] (Display section 17a) As the display unit 17a (see FIG. 7), for example, a device that displays a screen, such as a liquid crystal display (LCD), a plasma display panel (PDP), or an organic electroluminescence (EL) display, is used.

[0062] (Mobile 18) The mobile object 18 is equipped with a terminal 17. The mobile object 18 is, for example, a vehicle such as an automobile, a motorcycle, a bicycle, or a snowmobile, or an airplane or a ship. The mobile object 18 may also be a person carrying the terminal 17. In this embodiment, the mobile object 18 is an automobile.

[0063] (Communications Network 19) The communication network 19 is, for example, a communication network of the Internet, and is also constructed by a mobile phone line network, a wireless communication path, Ethernet (registered trademark), and the like.

[0064] (Mobile Information 24) The mobile object information 24 is information indicating the state of the mobile object 18. The mobile object information 24 is transmitted to the server 11 from the terminal 17, a drive recorder, or a navigation device.

[0065] [3. Hardware configuration] Next, the hardware configuration of each of the servers 11, 12, 13, 14, 15, 16 and the terminal 17 will be described with reference to FIGS.

[0066] (Hardware configuration of servers 11, 12, 13, 14, 15, 16) We will now explain the hardware configuration of the server 11. The hardware configurations of the servers 12, 13, 14, 15, and 16 are almost the same as that of the server 11, so their explanations will be omitted.

[0067] As shown in FIG. 6, the server 11 of this embodiment uses, for example, a computer. The server 11 includes a CPU (or GPU) 30. The CPU 30 is connected to, for example, a memory (hereinafter, referred to as a storage unit) 31, a connection port 33 for connecting / reading a storage device 32, and a communication circuit 34 for communicating with the outside via a network, via a bus line 35. The storage unit 31 stores a program 36 (36a) for processing the system 10. A browser program 37 and an OS 38 (operating system) may also be stored. The program 36 (36a) is installed in the server 11 by the storage device 32.

[0068] (Hardware configuration of terminal 17) In the present embodiment, the terminal 17 is, for example, a computer. As shown in Fig. 7, a display unit 17a is connected to a bus line 35 of the terminal 17. A terminal program 39 for processing the system 10 is stored in a storage unit 31 of the computer of the terminal 17. A browser program 37 and an OS 38 (operating system) may also be stored. The program 39 is installed in the terminal 17 by a storage device 32.

[0069] In this embodiment, the program 36 (36a) and the terminal program 39 may operate in cooperation with each other by utilizing the functions of the OS 38 and the browser program 37, respectively. Note that the program 36 (36a) and the terminal program 39 may operate independently without utilizing the browser program 37 and the OS 38, respectively.

[0070] In the hardware configuration of the above-mentioned program 36 (36a) and terminal program 39, the functions shown in the function book diagram of FIG. 2 are realized, for example, by using the CPU 30, program 36 (36a), and terminal program 39, but some or all of them may be sequence-controlled using a logic circuit such as a microcomputer, or a PLC (programmable logic controller).

[0071] [4. Program] (Flowchart showing the processing of the system 10) 8 is a flow chart showing an embodiment of the processing of the programs 36a, 36, 39 used respectively in the decision server 14, the server 11 and the terminal 17 in the system 10. The flow chart shows a content selection method 10c.

[0072] (S1: Acquisition of static and dynamic characteristic information) The CPU 30 of the decision server 14 (see FIG. 6) acquires the static property information 12a and the dynamic property information 13a.

[0073] (S2: Determination of area characteristic information 14a) The area characteristic information 14a is determined based on the static characteristic information 12a and the dynamic characteristic information 13a. The area characteristic information 14a is organized in the area characteristic DB 22 for each divided area 9.

[0074] (Area Characteristics DB22) 9 is a schematic diagram showing an example of the data structure of the area characteristic DB 22. The area characteristic DB 22 shown in the figure consists of area characteristic keywords 20a, emotion characteristic keywords 20b, and hot words 21a corresponding to a section area ID 9a. In this embodiment, static characteristic information 12a and dynamic characteristic information 13a are integrated.

[0075] (Others in Area Characteristics Information 14a) In acquiring the area characteristic information 14a, the area characteristic keywords 20a, the emotion characteristic keywords 20b, and the hot words 21a may be vectorized, and an average vector may be calculated to obtain the area characteristic information 14a. Alternatively, one or more words close to the average vector may be acquired to obtain the area characteristic information 14a. Note that the ratio of the number of keywords may be made different between the static characteristic information 12a and the dynamic characteristic information 13a, and the weights in the area characteristic information 14a may be changed during vectorization.

[0076] (Summary of Area Characteristics Information 14a) The area characteristic information 14a is acquired based on the static characteristic information 12a and the dynamic characteristic information 13a. That is, the static characteristic information 12a embodies the characteristics of an area consisting of geography and features that have been established in a specific region (area) for a relatively long time span and the geography and features of the area that will be established in the future, and incorporates dynamic characteristic information 13a that is in the "season" so to speak, such as a movement / trend that is a topic of conversation for a limited period of time related to the area, and is a concept that includes values / data that can be converted into them.

[0077] (S3: Sending location information 1a) Returning to FIG. 8, the location information 1 a is transmitted to the server 11 from the terminal 17 or a drive recorder of the automobile 18 .

[0078] (S4: Acquisition of location information 1a) The server 11 acquires the location information 1a.

[0079] (S5: Obtain area characteristic information 14a corresponding to location information 1a) The area characteristic information 14a of the section area 9 corresponding to the location information 1a is obtained from the decision server 14. In addition, the keyword 15b (see FIG. 5) is obtained from the content storage unit 15.

[0080] (S6: Matching) The area characteristic keywords 20a, emotion characteristic keywords 20b, and hot words 21a in the area characteristic information 14a are matched with the keywords 15b in the content list 15a.

[0081] In this embodiment, natural language processing is used for matching. For example, the area characteristic keywords 20a, the emotion characteristic keywords 20b, and the hot words 21a are each vectorized, and an average vector is obtained. On the other hand, the keywords 15b are vectorized. When there are multiple keywords 15b, each is vectorized, and an average vector is calculated. The similarity between the average vector of the area characteristic information 14a and the vector / average vector of the keywords 15b is calculated. The content list 15a of the keyword 15b with the highest similarity is selected.

[0082] (cosine analogue) In this embodiment, cosine similarity is used to measure the similarity of words or sentences. The cosine of two vectors is calculated, and the cosine value is used as the similarity. Cosine similarity indicates the closeness of the angle between the vectors, and the closer it is to 1, the more similar they are. The closer it is to 0, the less similar they are.

[0083] (Other matching) The degree of similarity between the area characteristic information 14a and the content list 15a may be determined by obtaining co-occurring words from the area characteristic keywords 20a, the emotion characteristic keywords 20b, and the hot words 21a, and by obtaining co-occurring words from the keywords 15b, and determining that the more overlapping words there are between the two, the higher the degree of similarity.Also, the degree of similarity between the two co-occurring words may be calculated by cosine similarity, and the content list 15a of the keyword 15b with the highest degree of similarity may be selected.

[0084] (S7: Extraction of content list 15a) The server 11 extracts the selected content list 15 a and stores it in the storage unit 31 .

[0085] (S8: Provision of content list 15a) The extracted content list 15 a is provided to the terminal 17 / vehicle 18 .

[0086] (S9: Acquiring extracted content list 15a) The extracted content list 15a is acquired by the terminal 17 / car 18.

[0087] 5. Other embodiments Next, a description will be given of other embodiments of the system 10. The modifications and other embodiments described below are substantially similar to the above-described system 10, so the same parts are given the same reference numerals and the description thereof will be omitted.

[0088] (System 10 Variation 1) We will now explain modified examples of the system 10. In modified example 1, the server 11 selects a content list 15a to be provided to the terminal 17, but does not extract the content. For this reason, the server 11 does not include an extraction unit 3. The content list 15a is provided to the terminal 17 from the content storage unit 15.

[0089] (System 10 Variation 2) A description will now be given of a variation of the system 10. In variation 2, the server 11 extracts the content list 15a to be provided to the terminal 17, but instead of transmitting it to the terminal 17 as is, the providing unit 4 transmits the reproduced data by streaming.

[0090] Second embodiment Next, another embodiment of the system 10 will be described. Returning to Fig. 2, in a system 10a according to the second embodiment, mobile object information 24 is transmitted from a terminal 17 / car 18 to a server 11. The server 11 determines the playback order of the extracted content list 15a based on the mobile object information 24 (see the description of the two-dot chain line in Fig. 8 (step S10)).

[0091] The mobile object information 24 is made up of parameters such as speed and the number of braking events in a given time. Points are set according to the parameter values. The parameter values ​​and points are organized in a mobile object information database (mobile object information DB) 25.

[0092] (Mobile Information DB25) FIG. 10 is a schematic diagram showing an example of the data structure of the mobile object information DB 25. In the mobile object information DB 25 shown in the figure, the reference numeral 26a indicates that the speed range is 0-10 km / h, the reference numeral 26b indicates that the speed range is 10-20 km / h, and so on. The reference numeral 27a indicates that the number of braking times is 0-5 times, the reference numeral 27b indicates that the number of braking times is 6-10 times, and the corresponding scores 25a are written in the columns below them. In this embodiment, elements that make driving smooth are added, and elements that hinder smooth driving are subtracted. Note that points may be added or subtracted based on other conditions. A score is obtained from each parameter of the current mobile unit information 24, and the sum of the scores is used to determine the score of the current mobile unit information 24.

[0093] (Content List Score Database 28) 11 is a schematic diagram showing an example of the data structure of the content list score database (content list score DB) 28. In the content list score DB 28 shown in the figure, reference numeral 28a indicates a content list ID, and reference numeral 28b indicates a score. The score of the content list 15a is composed of three axes, for example, "the drive is smooth," "not smooth," and "neither." It is also possible to set axes for elements such as the genre of the content and the weather.

[0094] The difference between the score of the current moving object information and the score of the content list is calculated, and the content list 15a is rearranged in ascending order of difference.

[0095] (Variation 3) A modification of the second embodiment 10a will be described. In the modification 3, the process of rearranging the content list 15a (see process S10) that determines the playback order of the content list 15a is executed by the terminal 17 / mobile body 18 instead of the server 11.

[0096] (Variation 4) A further modified example of the second embodiment 10a will be described. The server 11 of the modified example 4 acquires external information 16a from an external server 16. The external information 16a is information received from an external organization, such as weather, warning, traffic congestion information, and emergency earthquake alerts. For the external information, a score is calculated from a score table (not shown), and is added to or subtracted from the mobile object information 24 (see FIG. 10).

[0097] Third embodiment Next, a description will be given of still another embodiment of the system 10. Returning to Fig. 2, a system 10b according to a third embodiment includes a generation unit 5 that generates display elements 6 related to a content list 15a on a map screen 8 (see Fig. 13) of a display unit 17a displayed on a terminal 17 / mobile body 18.

[0098] (Generation part 5, display element 6) The generation unit 5 acquires keywords 15b from a content list 15a that are matched with the area characteristic information 14a for each divided area 9 on a map screen 8 of the display unit 17a. The acquired keywords 15b are subjected to sentiment analysis using, for example, a natural language processing model called BERT (Bidirectional Encoder Representations from Transformers). A trained model is used for the sentiment analysis. The result of the sentiment analysis is shown as a positive / negative score. Based on the acquired score, the font, character size, color, etc. are determined, and the display element 6 is generated. The scores, font, size, and color are organized, for example, in a display element database (display element DB) 7. Alternatively, the keywords 15b may be parameterized using other known techniques.

[0099] (Display element DB7) 12 is a schematic diagram showing an example of the data structure of a display element DB. The display element DB 7 shown in the figure consists of score 7a, font 7b, size 7c, and color 7d. Note that, instead of selecting large or small for positive / negative, the font may be made larger as the range of the corresponding section area 9 becomes larger.

[0100] (others) Another method of sentiment analysis is to use a Japanese evaluation polarity dictionary to obtain positive / negative scores.

[0101] (Map screen 8) FIG. 13 is a schematic diagram showing an example of a map screen 8 displayed on a terminal 17 / mobile body 18. In the figure, display elements 6 generated from keywords 15b in a content list corresponding to a section area are displayed. The display elements 6 are arranged on the map of the map screen 8 so as not to impede the visibility of the map. For example, they are positioned to avoid national highways and main roads (see symbol 6a: two-dot chain line). Note that while the figure shows the map screen 8 in a circular frame, it may also be shown in a rectangular frame.

[0102] Also, by arranging the display elements 6 along the roads, it is possible to prevent the display elements 6 from intersecting with the roads, which would impair the visibility of the map. "Along the road" may mean arranging the display elements 6 on the side of the road, or overlapping on top of the road. In the case of the road indicated by reference numeral 8a, although the display elements 6b are arranged on top of the road, they are arranged in the direction in which the road extends, which makes it possible to prevent the visibility of the map from being impaired.

[0103] Fig. 14 shows the map screen 8 in Fig. 13, with the point A enlarged. When the road runs north-south as shown by the reference symbol 8b, the display element 6c may be written vertically. When displaying in vertical writing, for example, when the road 8b to which the display element 6c is to be attached rises at a predetermined angle or more (rises in the north-south direction), the display element 6c may be displayed vertically. In this embodiment, for example, when the angle exceeds 45°, the display element 6c is displayed vertically.

[0104] (display position) The display positions of the display elements 6 generated by the generation unit 5 on the map screen 8 are processed by a display control unit 40 (see FIG. 2). The display control unit 40 reads, for example, the positions of roads in the map information 8, and then determines the display positions of the display elements 6. For example, in a section area 9 on a display screen 8, if there are few roads to guide / display to the user and there is space to display the display element 6, the display element 6a (see the two-dot dashed line in Figure 13) is displayed in a position that does not overlap with the roads, in other words, in the blank area. Also, when there are many roads to be guided / displayed to the user and the blank area for displaying the display elements 6 is small, the display elements 6b are displayed along the roads. Furthermore, when there are too many roads to guide / display to the user and displaying the display elements 6 would be cumbersome, the display elements 6 are not displayed. The display elements 6 may be displayed when blank areas are generated on the screen after the map screen 8 is enlarged or reduced. The display control unit 40 may be provided in the terminal 17 instead of the server 11.

[0105] [6.Other] The items described as "other" in the above-described embodiment can be used in appropriate combination. The functions of the external servers 12, 13, 14, 15, and 16 may be integrated into one or two external servers, or one or more external servers may be integrated into the server 11.

[0106] [7. Summary] (1) The system 10 (server 11, program 36, content selection method 10c) is characterized by comprising a content memory unit 15 that stores a content list 15a associated with keywords 15b, a static characteristic memory unit 12 that stores static characteristic information 12a including geographic information of a divided area 9 obtained by dividing a map into a plurality of areas, a location information acquisition unit 1 that acquires location information of a terminal 17, a determination unit 29 that determines area characteristic information 14a based on the static characteristic information of the divided area corresponding to the location information 1a and dynamic characteristic information 13a including hotwords 21a based on posted SNS information related to the corresponding divided area, a selection unit 2 that selects a corresponding content list based on the area characteristic information and the keywords, and a provision unit 4 that provides the selected content list to the terminal.

[0107] For this reason, static characteristic information that embodies the characteristics of an area, such as the geography and features that have been established in a specific region (area) over a relatively long time interval and the geography and features of the area that are likely to be established in the future, dynamic characteristic information that is, so to speak, "in season," such as movements / trends that are a topic of conversation for a limited period of time related to the area, and a content list of music can be matched, for example by language analysis, and a content list that takes into account the regional characteristics and trends in the specific area can be provided to people / vehicles, etc. located in the area.

[0108] (2) In such a system 10, the selection unit 2 is further based on the mobile object information 24 including the status of the mobile object 18 holding the terminal 17, so that it is possible to provide a content list that matches the status of the user holding the terminal 17. For example, if the content list 15a is selected with a priority, the playback order can be changed.

[0109] (3) In addition, the device is provided with a generation unit 5 that generates display elements 6 related to the content list 15a for display on a map screen 8 displayed on the terminal 17. This makes it possible to visualize display information on the map screen 8, allowing the user to enjoy traveling.

[0110] (4) The server of the present invention is characterized in that a map is divided into a plurality of areas, a content list 15a related to a specific divided area is set, and the server is provided with a generation unit 5 that generates display elements 6 related to the content list 15a for display in the divided area of ​​a map screen 8 displayed on a terminal 17. Therefore, the display elements can be visualized on the map screen 8, so that the user can enjoy moving around. The user can imagine the section area in which the display elements 6 are displayed. The user can imagine the music that plays when the user moves around the section area 9.

[0111] (5) In such a server, the display elements 6 are arranged so as to follow the roads 8a displayed on the map screen 8, thereby preventing the visibility of the map from being impaired. Also, since the display elements 6a are arranged so as to follow the roads 8a, it is possible to imagine the music that will be played as one travels along the roads 8a. [Explanation of symbols]

[0112] 1 Location information acquisition section 1a Location information 2 Selection section 3 Extraction part 4 Providing Department 5 Generation part 6 Display elements 6a Display elements 6b Display elements 6c Display elements 7 Display Element Database (Display Element DB) 7a score 7b font 7c Size 7d colors 8. Map screen 8a road 8b road 9. Area 9a Area ID 10 Content supply system (system) 10a Content supply system (system) 10b Content supply system (system) 10c Content Selection Method 11 Server 12 Static characteristic storage unit (server) 12a Static characteristic information 13 Dynamic characteristics memory unit (server) 13a Dynamic characteristics information 14 Decision Server (Server) 14a Area Characteristics Information 15 Content storage unit (server) 15a Content List 15b Keywords 16 External device (server) 16a External information 17 Terminals 17a Display section 18 Mobile vehicles (automobiles) 19. Communications Networks 20 Static Characteristics Database (Static Characteristics DB) 20a Area characteristic keywords 20b Emotional trait keywords 21 Dynamic Characteristics Database (Dynamic Characteristics DB) 21a Hot Words 22 Area Characteristics Information Database (Area Characteristics DB) 23 Content List Database (Content List DB) 24 Mobile Information 25 Mobile Information Database (Mobile Information DB) 25a Score 26a, 26b speed range 27a, 27b Number of brakes 28 Content List Score Database (Content List Score DB) 28a Content List ID 28b Scoring 29 Decision Section 30 CPU 31 Memory 32 Recording Devices 33 Connection Ports 34 Communication Circuits 35 Bus Line 36 Programs 37 Browser Programs 38 OS 39 Terminal Programs 40 Display control unit

Claims

1. A content storage unit that stores a list of content to which keywords are associated, A static characteristic storage unit that stores static characteristic information including geographic information of partitioned areas obtained by dividing a map into multiple areas, A location information acquisition unit that acquires the location information of the terminal, A determination unit that determines area characteristic information based on static characteristic information of the partitioned area corresponding to the location information and dynamic characteristic information including hot words based on posted SNS information relating to the corresponding partitioned area, A selection unit that selects the corresponding content list based on the area characteristic information and the keyword, The system includes a provisioning unit that provides a selected content list to the terminal, Content delivery system.

2. The content provision system according to claim 1, wherein the selection unit further selects the content list based on mobile information including the status of the mobile body holding the terminal.

3. The content provision system according to claim 1 or 2, further comprising a generation unit that generates display elements related to the content list for display on a map screen shown on the terminal.

Images