An artificial intelligence information feedback system
Through the artificial intelligence information feedback system, the use of popularity, keywords and related information subsystems to screen and output information, the problem of difficulty in obtaining information in smart homes is solved, the efficient and timed output of information is achieved, and the user's home information acquisition experience is improved .
Patent Information
- Application Number
- CN202010388888.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-05-10
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2040-05-10
AI Technical Summary
In smart homes, when users obtain information, due to the large amount of information and different terminals, it makes it difficult to screen information, waste time and energy, lack a regular and quantitative feedback system, and cannot effectively utilize home time.
An artificial intelligence information feedback system was designed, including a popularity information subsystem, a keyword information subsystem, an associated information subsystem, a local information database and a local collaborative subsystem. Information is obtained and filtered through these subsystems, and feedback trigger module is used. , information retrieval module and intelligent output device generate priority values according to user habits and preferences to realize intelligent output of information.
It effectively reduces the time and energy required for users to obtain information, and provides relevant information within the time period allowed by users through intelligent output devices, improving the efficiency and accuracy of information acquisition.
Smart Images

Figure CN111586180B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an information collection and feedback system, and more specifically, to an artificial intelligence information feedback system. Background Art
[0002] Smart home is the embodiment of Internet of Things under the influence of the Internet. Smart home connects various devices in the home (such as audio and video equipment, lighting system, curtain control, air conditioning control, security system, digital cinema system, audio and video server, video cabinet system, network home appliances, etc.) through the Internet of Things technology, providing home appliance control, lighting control, telephone remote control, indoor and outdoor remote control, anti-theft alarm, environmental monitoring, HVAC control, infrared forwarding and programmable timing control and other functions and means.
[0003] With the popularization of smart homes, users have more and more channels to obtain information, but the increase in the amount of information makes it impossible for users to filter the information. Users use their spare time to obtain information. Due to the different terminals and applications for obtaining information, users will waste more energy and time to get the information they want. There is no specific system for timely and quantitative feedback of information for users to use when they are at home. Summary of the invention
[0004] In view of this, an object of the present invention is to provide an artificial intelligence information feedback system.
[0005] In order to solve the above technical problems, the technical solution of the present invention is: an artificial intelligence information feedback system, including a heat information subsystem, a keyword information subsystem, a related information subsystem, a local information database and a local collaboration subsystem;
[0006] The popularity information subsystem is used to obtain first information from an external popularity information database, the first information is obtained in popularity order and forms popularity information and is sent to the local collaboration subsystem, and the popularity information database obtains information through a news platform;
[0007] The keyword information subsystem is connected to an external keyword information database to obtain second information, the second information is obtained in a matching manner according to the local keywords received from the local collaboration subsystem, and the keyword information database obtains information through a platform subscribed by the user;
[0008] The associated information subsystem is used to obtain third information from an external associated information database, the third information is obtained in a matching manner according to the user browsing habits generated by the local collaborative subsystem, and the associated information database obtains information through all platforms connected to the artificial intelligence information feedback system;
[0009] The local information database is used to receive the first information, the second information and the third information as local information;
[0010] The local collaborative subsystem is configured with a feedback trigger module, an information retrieval module, a local information management module and an intelligent output device;
[0011] The feedback trigger module is used to generate a trigger request and send the trigger request to the information retrieval module, wherein the trigger request includes a trigger period and an output device number;
[0012] The local information management module stores basic user information, and the local information management module is configured with a preference management strategy, the preference management strategy is used to manage the priority value of each local information, and the preference management strategy adjusts the priority value of each local information according to the basic user information;
[0013] The information retrieval module is configured with an information retrieval strategy, which retrieves data from the local information database according to a trigger request to generate an output queue, and outputs the corresponding local information in the order of the output queue through the corresponding intelligent output device. The information retrieval strategy determines the local information of the output queue according to the information priority value in the local information database so that the sum of the priority values of the local information within the trigger period is maximized.
[0014] Furthermore, the local information management module is configured with a thermal management strategy, and the thermal management strategy is configured with a first attenuation number, a second attenuation number and a third attenuation number, and the thermal management strategy reduces the priority value of local information corresponding to the first information every first preset time according to the first attenuation number; the thermal management strategy reduces the priority value of local information corresponding to the second information every second preset time according to the second attenuation number, and the thermal management strategy reduces the priority value of local information corresponding to the third information every third preset time according to the third attenuation number.
[0015] Furthermore, the heat information subsystem is configured with a heat number, and the heat information subsystem retrieves the corresponding first information according to the heat number, and the heat number is generated according to the time length corresponding to the trigger period.
[0016] Furthermore, the keyword information subsystem is configured with keywords, and the keyword information subsystem retrieves corresponding second information according to the keywords, and the keywords are generated according to the identity information of the user.
[0017] Furthermore, the keyword information subsystem is configured with keywords, and the keyword information subsystem retrieves corresponding second information according to the keywords, and the keywords are generated according to the user's preference information.
[0018] Furthermore, the keyword information subsystem is configured with keywords, and the keyword information subsystem retrieves corresponding second information according to the keywords, and the keywords are generated according to the user's status information.
[0019] Furthermore, the user browsing habit is generated based on the second information.
[0020] Furthermore, the basic user information includes identity information, preference information and status information.
[0021] Furthermore, the preference management strategy includes extracting a basic feature model based on user basic information, generating an associated feature extraction strategy based on the basic information model, determining the degree of association between each local information and the basic feature model based on the associated feature extraction strategy, and generating the priority value based on the degree of association.
[0022] Furthermore, the basic user information is set through user input.
[0023] The technical effects of the present invention are mainly reflected in the following aspects: through such settings, corresponding information is obtained according to user habits and basic conditions, and then output through the output device in the corresponding smart home system within the time period allowed by the user, avoiding the waste of time and energy required for users to find information. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0025] Figure 1 : Schematic diagram of the system architecture of the present invention;
[0026] Figure 2 : Schematic diagram of the local collaborative subsystem of the present invention.
[0027] Attached figure numbers: 100, popularity information subsystem; 200, keyword information subsystem; 300, related information subsystem; 400, local information database; 500, local collaboration subsystem; 510, feedback trigger module; 520, information retrieval module; 530, local information management module; 540, intelligent output device. DETAILED DESCRIPTION
[0028] The specific implementation modes of the present invention are further described below in conjunction with the accompanying drawings to make the technical solutions of the present invention easier to understand and grasp.
[0029] An artificial intelligence information feedback system includes a heat information subsystem 100, a keyword information subsystem 200, a related information subsystem 300, a local information database 400 and a local collaboration subsystem 500;
[0030] The heat information subsystem 100 is used to obtain first information from an external heat information database. The first information is obtained in order of heat and forms heat information and is sent to the local collaborative subsystem 500. The heat information database obtains information through a news platform; the heat information subsystem 100 is configured through a server and then obtains information by connecting to a news platform. Since the information heat of news platforms is different, higher heat information can be obtained according to the news platform. Preferably, the news platform will classify news, such as technology, politics, entertainment, etc. Users can pre-specify the heat preference value of the type according to this classification. The heat value multiplied by the heat is a specific heat value. For example, if the user is more interested in technology, then there will be more technology news in the first information. It can play a real-time and customer-required information output situation and can be customized according to customer needs.
[0031] The keyword information subsystem 200 is connected to an external keyword information database to obtain the second information. The second information is obtained in a matching manner based on the local keywords received from the local collaboration subsystem 500. The keyword information database obtains information through the platform subscribed by the user. There are two sources of keywords. The first one can be set according to demand. The user can customize the keywords of interest. Then the system will collect this information in advance and wait for the user's free time to output it. These keyword information is obtained by matching, not limited to completely identical keywords. The semantic analysis algorithm can be nested and the system can retrieve the results with a high matching degree to give the user a recommendation. The second one can be obtained through the user's basic information. For example, if the user's identity is a surgeon, then the keyword can be generated by the keyword under the surgeon's list. The difference from the first information is that this is generated according to the user's subscription action. Since the information on the Internet is relatively complex, for example, the user subscribes to a medical platform-related database, and then according to the keywords of surgical operations, it is easier to find relevant data from this database. The keyword information database is updated according to the subscription platform and updated in real time to ensure the source and reliability of the data. The keyword information subsystem 200 is configured with keywords. The keyword information subsystem 200 retrieves the corresponding second information according to the keywords, and the keywords are generated according to the user's identity information. The keyword information subsystem 200 is configured with keywords, and the keyword information subsystem 200 retrieves corresponding second information according to the keywords, and the keywords are generated according to the user's preference information. The keyword information subsystem 200 is configured with keywords, and the keyword information subsystem 200 retrieves corresponding second information according to the keywords, and the keywords are generated according to the user's status information.
[0032] The associated information subsystem 300 is used to obtain the third information from the external associated information database. The third information is obtained in a matching manner based on the user browsing habits generated by the local collaborative subsystem 500. The associated information database obtains information through all platforms connected to the artificial intelligence information feedback system; the user's browsing habits are generated according to the following logic: one is to obtain tracking information by tracking the browsing traces of the account according to the user's registered account, and then a more reasonable information tracking effect can be obtained, and then the browsing habits can be generated by matching through the traces, that is, according to this logic, the browsing habits that the user needs more can be generated, so that the data that can be used as the third information can be determined. The second is that the user's browsing habits can be determined according to the selection frequency of the second information. For example, in the information where the second information is selected more, the same or corresponding information in the selected data is obtained through a data cross algorithm to form a basis for determining the user's browsing habits. For example, the frequency of the word "sports" in the second information is high, which means that the user has a preference for this information or the user situation pushes information with this keyword for better effect. The basic information of the user is set by user input. According to the user's real-time update, the user's situation can be obtained to customize the optimal information push strategy, and the user can also set the priority value ratio corresponding to different types. For example, a higher priority value can be set for preferences, and the priority value related to preferences is higher. In the preferred embodiment of the present invention, the priority value is formed according to the model. For example, the priority value is: (A1a1+A2a2K+A n a n )+(B1b1+B2b2K+B n b n )+(C1c1+C2c2K+C n c n ) For example, A is the priority value cardinality corresponding to the Nth identity information, which is set according to user needs, and B corresponds to the priority value cardinality corresponding to the Nth preference information, and C corresponds to the priority value cardinality corresponding to the Nth status information. The corresponding lowercase letters are preset weights and proportions. In this way, the priority value of each local information can be determined based on this model, and the determination method is based on the core words in the information, the corresponding categories, and the number of times the keywords appear. Each local information may involve relevance in multiple directions, so the priority value is also calculated in multiple dimensions under this model to obtain an optimal result.
[0033] The local information database 400 is used to receive the first information, the second information and the third information as local information; and the local information database 400 is used to make a basis for the next screening. The local information is used to achieve information security and reliability, ensure the rationality of the data and improve the use effect of the data. At the same time, it ensures that the local storage resources are within a reasonable range.
[0034] The local collaborative subsystem 500 is configured with a feedback trigger module 510, an information retrieval module 520, a local information management module 530 and an intelligent output device 540; first of all, the intelligent output device 540, which is set in the home Internet of Things, can be a home video and audio output device, or a mobile device. Different intelligent output devices 540 are selected, and the output method is also different, so the corresponding information is also different. For example, voice output, text output, and image output information need to be coordinated according to user needs, and the output time length and time period are also important bases for the content of the output information.
[0035] The feedback trigger module 510 is used to generate a trigger request and send the trigger request to the information retrieval module 520. The trigger request includes a trigger period and an output device number. First, the feedback trigger module 510 can generate a trigger request based on actual security and reliability. The trigger request can be generated by user input or by actual occurrence. Through such actual occurrence, the actual usage time of the user can be determined, as well as the appropriate time to start outputting information. The preferred information format can be determined based on the device number, and format conversion or information screening can be performed in advance.
[0036] The local information management module 530 stores basic user information. The local information management module 530 is configured with a preference management strategy. The preference management strategy is used to manage the priority value of each local information. The preference management strategy adjusts the priority value of each local information according to the basic user information. The preference management strategy is used to manage the priority value of local information. In this way, the priority value of each local information can be used as the basis for information output to ensure the security and reliability of data. The priority value reflects the user's priority for different information. The basic user information includes identity information, preference information and status information. For example, the user's occupation and family relationship. The occupation can push information related to the employment field, and the family relationship can push information such as education and caring for the elderly. Preference information such as user hobbies, such as the user's hobbies of comics and singing, then these information can be summarized accordingly to recommend relevant information to the user, such as the user's physical condition or recent condition, etc., can output corresponding information, for example, the user is in a state of exercise recently, then relevant information can be pushed, and it can play a better role in collecting and obtaining information. The preference management strategy includes extracting a basic feature model based on user basic information, generating an associated feature extraction strategy based on the basic information model, determining the degree of association between each local information and the basic feature model based on the associated feature extraction strategy, and generating the priority value based on the degree of association.
[0037] The information retrieval module 520 is configured with an information retrieval strategy, which retrieves data from the local information database 400 according to the trigger request to generate an output queue, and outputs the corresponding local information in the order of the output queue through the corresponding intelligent output device 540. The information retrieval strategy determines the local information of the output queue according to the information priority value in the local information database 400, so that the sum of the priority values of the local information in the trigger period is maximized. This solution is to output the corresponding information within the time period preset or permitted by the user according to the importance of different information to the user, and the most important information is output in time order, and then output in sequence, so that the user can obtain information from different platforms in a concentrated idle time period, and this information is output according to the device used by the user. Since the time length occupied by each information is different, the core of the present invention is to set an algorithm for time and priority, calculate and obtain the combination with the largest priority value in a limited time period as output, not only considering the priority value but also allowing the user to obtain more and more important information in a limited time, and continuously training this priority association algorithm can achieve a better association effect.
[0038] The local information management module 530 is configured with a heat management strategy, which is configured with a first attenuation number, a second attenuation number, and a third attenuation number. The heat management strategy reduces the priority value of the local information corresponding to the first information every first preset time according to the first attenuation number; the heat management strategy reduces the priority value of the local information corresponding to the second information every second preset time according to the second attenuation number; the heat management strategy reduces the priority value of the local information corresponding to the third information every third preset time according to the third attenuation number. The first attenuation number is greater than the second attenuation number, and the second attenuation number is greater than the third attenuation number. Because news information is more likely to decay due to time, a higher attenuation number can be set, which means that the priority value of this type of information decreases faster. The second attenuation number corresponding to the second information can be based on user subscriptions, so that the learning section may also be a report in a certain field or a weather forecast, etc., so these information can also be set with an attenuation number. The third type is information recommended according to user habits. This type of information may also be verified to be accurate relative to time. When the priority value is 0, the corresponding data will be deleted from the database to ensure data load.
[0039] The heat information subsystem 100 is configured with a heat number, and the heat information subsystem 100 retrieves the corresponding first information according to the heat number, and the heat number is generated according to the time length corresponding to the trigger period.
[0040] Of course, the above are only typical examples of the present invention. In addition, the present invention may also have many other specific implementations. All technical solutions formed by equivalent replacement or equivalent transformation fall within the scope of protection required by the present invention.
Claims
1. An artificial intelligence information feedback system, Features: It includes the popularity information subsystem, keyword information subsystem, related information subsystem, local information database and local collaboration subsystem; The popularity information subsystem is used to obtain first information from an external popularity information database, the first information is obtained in popularity order and forms popularity information and is sent to the local collaboration subsystem, and the popularity information database obtains information through a news platform; The keyword information subsystem is connected to an external keyword information database to obtain second information, the second information is obtained in a matching manner according to the local keywords received from the local collaboration subsystem, and the keyword information database obtains information through a platform subscribed by the user; There are two sources of keywords. The first one is set according to demand. Users customize the keywords they are interested in. Then the system will collect this information in advance and wait for the user's free time to output it. These keyword information is obtained through matching. The nested semantic analysis algorithm and the system retrieve the results with higher matching degree and give recommendations to users. The second one is obtained through the user's basic information. The difference from the first information is that this is generated according to the user's subscription action, while the keyword information library is updated according to the subscription platform and updated in real time. The keyword information subsystem is configured with keywords. The keyword information subsystem retrieves the corresponding second information according to the keywords. The keywords are generated according to the user's identity information. The keyword information subsystem is configured with keywords. The keyword information subsystem retrieves the corresponding second information according to the keywords. The keywords are generated according to the user's preference information; The keyword information subsystem is configured with a keyword, the keyword information subsystem retrieves corresponding second information according to the keyword, and the keyword is generated according to the user's status information; The associated information subsystem is used to obtain the third information from the external associated information database. The third information is obtained in a matching manner based on the user browsing habits generated by the local collaborative subsystem. The associated information database obtains information through all platforms connected to the artificial intelligence information feedback system. The user's browsing habits are generated according to the following logic. One is to obtain tracking information by tracking the browsing traces of the account according to the user's registered account, and then a more reasonable information tracking effect is obtained. Then, the browsing habits are generated by matching through the traces, that is, the browsing habits that the user needs more are generated according to this logic, so that the data used as the third information can be determined; The second is to determine the browsing habits of the user according to the selection frequency of the second information, including obtaining the same or corresponding information in the selected data through a data cross-algorithm from the information in which the second information is selected the most, forming a basis for determining the browsing habits of the user; the basic information of the user is set by the user input, and the user situation is obtained according to the real-time update of the user to customize the optimal information push strategy, and the user also sets the priority value ratio corresponding to different types, including setting a higher priority value for the preference, and the priority value related to the preference is higher. The priority value is formed according to the model, and the priority value is: (A 1 a 1 +A 2 a 2 …+A n a n )+(B 1 b 1 +B 2 b 2 …+B n b n )+(C 1 c 1 +C 2 c 2 …+C n c n ) A is the priority value cardinality corresponding to the Nth identity information, which is set according to user needs, while B corresponds to the priority value cardinality corresponding to the Nth preference information, and C corresponds to the priority value cardinality corresponding to the Nth status information. The corresponding lowercase letters are preset weights and proportions. The priority value of each local information is determined according to this model. The determination method is based on the core words in the information, the corresponding categories, and the number of times the keywords appear. Each local information will involve relevance in multiple directions. The priority value is also calculated in multiple dimensions under this model to obtain an optimal result. The local information database is used to receive the first information, the second information and the third information as local information; The local collaborative subsystem is configured with a feedback trigger module, an information retrieval module, a local information management module and an intelligent output device; The feedback trigger module is used to generate a trigger request and send the trigger request to the information retrieval module, wherein the trigger request includes a trigger period and an output device number; The trigger request is generated by user input or by actual occurrence. The actual usage time of the user is determined by such actual occurrence, and the appropriate time to start outputting information is determined. The information format is determined according to the device number, and the format conversion or information screening is performed in advance. The local information management module stores basic information of the user, and the local information management module is configured with a preference management strategy, the preference management strategy is used to manage the priority value of each local information, and the preference management strategy adjusts the priority value of each local information according to the basic information of the user; the priority value reflects the priority of the user to different information, and the basic information of the user includes identity information, preference information and status information; the preference management strategy includes extracting a basic feature model according to the basic information of the user, generating an associated feature extraction strategy according to the basic information model, determining the association degree of each local information with the basic feature model according to the associated feature extraction strategy, and generating the priority value according to the association degree; The information retrieval module is configured with an information retrieval strategy, which retrieves data from the local information database according to the trigger request to generate an output queue, and outputs the corresponding local information in the order of the output queue through the corresponding intelligent output device, and the information retrieval strategy determines the local information of the output queue according to the information priority value in the local information database, so that the sum of the priority values of the local information in the trigger period is maximized; The local information management module is configured with a heat management strategy, the heat management strategy is configured with a first attenuation number, a second attenuation number, and a third attenuation number, and the heat management strategy reduces the priority value of the local information corresponding to the first information every first preset time according to the first attenuation number; The thermal management strategy reduces the priority value of local information corresponding to the second information every second preset time according to the second attenuation number, and the thermal management strategy reduces the priority value of local information corresponding to the third information every third preset time according to the third attenuation number.
2. An artificial intelligence information feedback system as claimed in claim 1, Features: The heat information subsystem is configured with a heat number, and the heat information subsystem retrieves the corresponding first information according to the heat number, and the heat number is generated according to the time length corresponding to the trigger period.
