User state information acquisition method, apparatus, device, and medium
By classifying and extracting feature information from the target user's memory data, user status assessment text is generated, which solves the problem of inaccurate user status information acquisition in existing technologies and achieves automated and highly accurate user status information acquisition.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- PING AN TECH (SHENZHEN) CO LTD
- Filing Date
- 2023-07-17
- Publication Date
- 2026-06-05
AI Technical Summary
The accuracy of user status information acquisition in existing technologies is low, especially in cases of memory decline, where individual differences lead to insufficient accuracy due to the inability to obtain status information manually.
By acquiring the target user's first memory data, performing scene classification processing, extracting scene information and feature information, generating user status assessment text, and automatically obtaining user status information using the user status assessment text.
It improves the accuracy of user status information acquisition, can automatically generate user status assessment text, and enhances the intelligence and accuracy of status information acquisition.
Smart Images
Figure CN116994247B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of data processing technology and medical and health technology, and in particular to a method, apparatus, device and medium for obtaining user status information. Background Technology
[0002] With the development of the times, people's lifestyles and habits have undergone tremendous changes. Due to busy work schedules, people need to do more mental labor and may often work overtime late into the night, leading to memory decline, and in severe cases, even memory loss. For example, some elderly people's memories gradually fade with age. Therefore, in response to the above problems, people have developed many memory consolidation methods to help people solidify their memories.
[0003] Current solutions for addressing memory decline typically involve investing significant human and material resources in cognitive training, life skills training, and social activities. After addressing memory decline through these tasks, it's necessary to assess the user's state. This state information is usually obtained manually by the user who performed the tasks. However, due to individual differences, the accuracy of the obtained user state information is relatively low. Summary of the Invention
[0004] This invention provides a method, apparatus, device, and medium for acquiring user status information, in order to solve the technical problem of low accuracy of the acquired user status information.
[0005] Firstly, a method for obtaining user status information is provided, the method comprising:
[0006] Acquire the first memory data of the target user, the first memory data including a first image set, and the target user included in each first image in the first image set;
[0007] The first image set is subjected to scene classification processing to obtain K first image groups, and the scene information corresponding to each of the K first image groups is different;
[0008] Obtain scene information corresponding to each of the K first image groups to obtain a first scene information set;
[0009] Based on the first scene information set and the first images in the K first image groups, determine the scene feature information corresponding to each of the K first image groups;
[0010] Based on the scene feature information corresponding to each of the K first image groups, the user status evaluation text of the target user is determined;
[0011] The user status information of the target user is obtained based on the user status assessment text.
[0012] Secondly, a user status information acquisition device is provided, the user status information acquisition device comprising:
[0013] The first acquisition unit is used to acquire the first memory data of the target user, the first memory data including a first image set, and each first image in the first image set including the target user;
[0014] A classification unit is used to perform scene classification processing on the first image set to obtain K first image groups, and the scene information corresponding to each of the K first image groups is different;
[0015] The second acquisition unit is used to acquire scene information corresponding to each of the K first image groups to obtain a first scene information set;
[0016] The first determining unit is configured to determine scene feature information corresponding to each of the K first image groups based on the first scene information set and the first images in the K first image groups;
[0017] The second determining unit is used to determine the user status evaluation text of the target user based on the scene feature information corresponding to each of the K first image groups.
[0018] The third acquisition unit is used to acquire the user status information of the target user based on the user status evaluation text.
[0019] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described user status information acquisition method.
[0020] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described user status information acquisition method.
[0021] In the aforementioned scheme implemented by the user status information acquisition method, device, equipment, and medium, the first memory data of the target user can be acquired. This first memory data includes a first image set, where each first image in the first image set includes the target user. Scene classification processing is performed on the first image set to obtain K first image groups. The scene information corresponding to each of the K first image groups is different. Scene information corresponding to each of the K first image groups is acquired to obtain a first scene information set. Based on the first scene information set and the first images in the K first image groups, scene feature information corresponding to each of the K first image groups is determined. Based on the scene feature information corresponding to each of the K first image groups, scene feature information is determined. Based on the corresponding scene feature information, the user status assessment text of the target user is determined, and the user status information of the target user is obtained according to the user status assessment text. In this invention, the scene corresponding to the first memory data of the target user can be classified to obtain K first image groups. Scene feature information is extracted from the K first image groups to obtain scene feature description information corresponding to the K first image groups. Based on the scene feature description information corresponding to the K first image groups, a user status assessment text corresponding to the target user is formulated. Thus, user status information can be obtained by using the status assessment text, thereby automatically generating user status assessment text and obtaining the user status information of the target user based on the user status assessment text, improving the accuracy of user status information acquisition. Attached Figure Description
[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0023] Figure 1 This is a schematic diagram of an application environment for a user status information acquisition method according to an embodiment of the present invention;
[0024] Figure 2 A flowchart illustrating a method for obtaining user status information provided in an embodiment of this application;
[0025] Figure 3 This is a schematic diagram of the user status information acquisition device in one embodiment of the present invention;
[0026] Figure 4 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention; Detailed Implementation
[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0028] The user status information acquisition method provided in this embodiment of the invention can be applied to, for example... Figure 1 In this application environment, the client communicates with the server via a network. In healthcare scenarios, since target users may experience memory decline, it's necessary to acquire their user status information and perform status analysis to determine the extent of memory decline. When acquiring the target user's user status information, related users can upload the target user's primary memory data to the server via the client. The target user may be an elderly person experiencing memory decline, and related users may be the target user's relatives, doctors, nurses, etc. The primary memory data can be multimodal data such as text, voice recordings, images, and video recordings corresponding to scenes experienced by the target user. When the client acquires the target user's user status information, it can obtain this primary memory data from the server. The client can perform scene analysis on the initial memory data and generate corresponding user status assessment text to display to the user. This assessment text could be, for example, "What is there against the north wall in the living room?" or "What color is the TV, and where is it placed?" The client also receives user status information descriptions input by the target user based on these assessment texts. These descriptions could be, for example, "There is a white table against the north wall in the living room." This information is then analyzed to obtain the user's overall status information. Therefore, by obtaining the offset between historical user status information stored on the server and the target user's corresponding status information, the client can determine the target user's recovery status. This allows therapists to make appropriate adjustments to the target user's treatment tasks based on their recovery progress, thus improving the overall intelligence of the treatment.
[0029] The client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The invention will now be described in detail through specific embodiments.
[0030] Please see Figure 2 As shown, Figure 2A flowchart illustrating a user status information acquisition method provided in an embodiment of the present invention includes the following steps:
[0031] S201. Obtain the first memory data of the target user, the first memory data including a first image set, and each first image in the first image set includes the target user.
[0032] The primary memory data can be provided and uploaded to the server of the user status information acquisition device by the target user's associated users. Associated users can be the target user's relatives, treating physicians, treating nurses, etc. The primary memory data includes a primary image set, where the target user's primary memory data can be multimodal data such as text, voice recordings, images, and video recordings corresponding to the language descriptions of scenes experienced by the target user. The images in the primary image set can be images from the primary memory data, etc.
[0033] S202. Perform scene classification processing on the first image set to obtain K first image groups, each of the K first image groups having different scene information.
[0034] The classification can be based on the scene corresponding to each image in the first image set. Images with the same or similar scenes in the first image set are grouped together to obtain K first image groups. The scene information corresponding to each of the K first image groups is different.
[0035] For example, if the first image set contains images of the target user's home, the park the user frequently visits, the trade market, and the residential community where the target user lives, the first image set can be divided into four first image groups based on "home", "park", "trade market" and "residential community". This facilitates feature extraction from the first image groups, reduces the computational load of feature extraction, and improves the efficiency of obtaining user status information.
[0036] S203. Obtain scene information corresponding to each of the K first image groups to obtain a first scene information set.
[0037] Scene information is extracted from each image in the K first image groups to obtain first scene information corresponding to each of the K first image groups. This scene information is then defined as an element in a first scene information set to obtain the first scene information set. The first scene information can be a textual description of the scene corresponding to an image in a first image group. For example, if a first image group represents images of a target user's living room, and the living room contains a table, a chair, and a television, then the scene information corresponding to this first image group is: "There is a table, a chair, and a television in the living room."
[0038] S204. Based on the first scene information set and the first images in the K first image groups, determine the scene feature description information corresponding to each of the K first image groups.
[0039] It can be that elements are extracted from the set of first scene information corresponding to the first images in the K first image groups to obtain scene feature information corresponding to the first images in the K first image groups respectively, and description information corresponding to the scene feature information is obtained to obtain scene feature description information corresponding to each of the K first image groups respectively.
[0040] S205. Determine the user status evaluation text of the target user based on the scene feature description information corresponding to each of the K first image groups.
[0041] The process can involve obtaining K sub-user status evaluation texts based on scene feature description information corresponding to each of the K first image groups, and then obtaining the correlation information between every two sub-user status evaluation texts to obtain a first correlation information set. The K sub-user status evaluation texts are then fused based on this first correlation information set to obtain a user status evaluation text. This user status evaluation text can be used to obtain the user status information of a target user.
[0042] User status assessment text can be relevant questions formulated for the target user based on the obtained scene feature description information. For example, if the scene feature description information is "a white square table is placed in the middle of the north wall, a yellow round chair is placed 5 meters south of the white square table, and a black rectangular TV is placed in the center of the white square tabletop", then the user status assessment text could be: "What is there in the north wall of the living room?" or "What color is the TV and where is it placed?"
[0043] S206. Obtain the user status information of the target user based on the user status assessment text.
[0044] The system displays a user status assessment text to the target user. The target user can then describe their current user status based on the received assessment text and input this description into the client. The client analyzes this description to obtain the target user's specific user status information. This status information can represent the user's current state; for example, it could indicate declining memory or improved memory.
[0045] In one possible implementation, the accuracy of user status information acquisition is improved by showing user status assessment text to the target user and receiving user status information description information output by the target user in response to the user status assessment text, and then analyzing the obtained user status description information to obtain the user status information corresponding to the target user. Specifically, this method includes:
[0046] A1. Display the user status assessment text.
[0047] The process involves displaying a user status assessment text to the target user, prompting them to respond accordingly. The user's response to the assessment text is then used to define their user status description. This user status assessment text can be relevant questions tailored to the target user based on obtained scene feature descriptions. For example, if the scene feature description is "A white square table is placed in the middle against the north wall, a yellow round chair is placed 5 meters south of the white square table, and a black rectangular television is placed in the center of the white square tabletop," then the user status assessment text could be: "What is there against the north wall in the living room?" or "What color is the television, and where is it placed?"
[0048] A2. Receive user status description information from the target user in the text input of the user status evaluation.
[0049] The user status description information can be the target user's response to the user status assessment text. For example, if the user status assessment text is "What is there against the north wall in the living room?", then the user status description information could be "There is a white table against the north wall in the living room".
[0050] A3. Determine the user status information of the target user based on the user status description information.
[0051] The user status description information corresponding to the target user is segmented, semantically extracted, and information fused to obtain the user status information corresponding to the target user.
[0052] In this example, by displaying a user status assessment text corresponding to the target user and prompting the target user to respond accordingly, the user status description information is obtained. Analyzing this description information allows for accurate acquisition of the target user's status information, thus improving the accuracy of user status information retrieval.
[0053] In one possible implementation, the user state description information is segmented into sentences, and semantic and keyword extraction is performed on the segmented user state description information to obtain first semantic information and a first keyword set. Based on the first semantic information and the first keyword set, first reference user state information and second reference user state information are determined. The first reference user state information and the second reference user state information are then fused to obtain the user state information. The specific operation steps are as follows:
[0054] B1. The user status description information is segmented to obtain M first statements.
[0055] Retrieve the delimiters in the user status description information, including periods, semicolons, question marks, exclamation marks, etc., and divide the user status description information into M first statements based on the delimiters.
[0056] B2. Obtain the first semantic information corresponding to each of the M statements.
[0057] Semantic analysis is performed on M first statements to obtain the first semantic information corresponding to each of the M first statements. This semantic analysis can be performed on the M first statements using a general semantic analysis method.
[0058] B3. Determine the first reference user status information of the target user based on the first semantic information corresponding to the M first statements respectively.
[0059] It can be that the first semantic information corresponding to each of the M first statements is vector encoded to obtain the vector representation of the first semantic information corresponding to each of the M first statements, the vector representations of the first semantic information corresponding to each of the M first statements are vector fused to obtain the vector representation of the first reference user state information, and the vector representation of the first reference user state information is decoded to obtain the first reference user state information.
[0060] B4. Extract keywords from each of the M first statements to obtain the first keywords corresponding to each of the M first statements. Determine the first keywords corresponding to each of the M first statements as elements in the first keyword set to obtain the first keyword set.
[0061] One can extract keywords from each of the M first statements using common keyword extraction methods to obtain the first keywords corresponding to each of the M first statements.
[0062] B5. Based on the set of first keywords corresponding to the M first statements, determine the second reference user status information of the target user.
[0063] The knowledge representation corresponding to the first keyword in the first keyword set is obtained, and knowledge fusion is performed on the knowledge representation corresponding to the first keyword in the first keyword set to obtain the second reference user state information of the target user. The knowledge fusion of the knowledge representation corresponding to the first keyword in the first keyword set can be performed using a general knowledge fusion method. For example, each first keyword in the first keyword set is vector-encoded to obtain a keyword vector corresponding to each first keyword in the first keyword set, and the keyword vectors corresponding to each first keyword in the first keyword set are fused according to the triangle rule in vector operations.
[0064] B6. The first reference user status information and the second reference user status information are fused to obtain the user status information of the target user.
[0065] The process can involve vector encoding the first reference user state information and the second reference user state information to obtain the first reference user state information vector and the second reference user state information vector, vector fusion of the first reference user state information vector and the second reference user state information vector to obtain the target user's user state information vector, and decoding the target user's user state information vector to obtain the target user's user state information.
[0066] In this example, the user status description information is segmented into sentences, and semantic and keyword extraction is performed on the segmented user status description information to obtain first reference user status information and second user status information. The first reference user status information and the second reference user status information are then fused to obtain the target user's user status information. This allows the first reference user status information and the second reference user status information to be cross-verified, improving the accuracy of user status information acquisition.
[0067] In one possible implementation, the first reference user state information and the second reference user information are vector-encoded to obtain a first reference user state information vector and a second reference user information vector. These vectors are then fused to obtain a target user state information vector. Finally, the target user state information vector is decoded to obtain the target user's user state information. The specific steps are as follows:
[0068] C1. Encode the first reference user state information to obtain the first reference user state information vector.
[0069] One approach is to use word vectorization to encode the state information of the first reference user, thereby obtaining a vector representing the state information of the first reference user. Word vectorization refers to a method that maps words or phrases to vectors in a low-dimensional space. Through vector computation, semantic analysis or knowledge fusion can be performed on multiple words or phrases.
[0070] C2. Encode the second reference user state information to obtain the second reference user state information vector.
[0071] One approach is to use word vectorization to encode the second reference user's state information to obtain a second reference user state information vector.
[0072] C3. Perform vector fusion on the first reference user state information vector and the second reference user state information vector to obtain the target user state information vector.
[0073] This can be achieved by mapping the first reference user state information vector to the low-dimensional space where the second reference user state information vector resides, translating the first reference user state information vector so that the first reference user state information vector and the second reference user state information vector have a common starting point, and using the triangle rule of vectors to perform calculations to achieve the purpose of vector fusion of the first reference user state information vector and the second reference user state information vector, so as to obtain the target user state information vector.
[0074] C4. Determine the user status information of the target user based on the target user status information vector.
[0075] The target user state information vector is decoded, which can be done by using the inverse process of word vector encoding to obtain the target user state information.
[0076] In this example, the first and second reference user state information are vector-encoded, enabling mathematical calculations to be performed on them. This allows for the fusion of the first and second reference user state information, thereby accurately obtaining the user state information corresponding to the target user. This improves the accuracy of user state information acquisition.
[0077] In one possible implementation, scene analysis is performed on each of the K first image groups to obtain scene feature description information corresponding to each of the K first image groups. Feature extraction is then performed on the scene feature description information corresponding to each of the K first image groups to obtain scene feature information corresponding to each of the K first image groups. The specific operation steps are as follows:
[0078] D1. Based on the first scene information set, determine the scene feature information corresponding to each of the K first image groups.
[0079] It can be that elements are extracted from the first scene corresponding to the K first image groups in the first scene information set to obtain scene feature information corresponding to the K first image groups respectively.
[0080] For example, if a scene in the first scene information set is "There is a table, a chair and a television in the living room", and the elements extracted from this scene information are "table", "chair" and "television", then the scene feature information corresponding to "There is a table, a chair and a television in the living room" is "table", "chair" and "television".
[0081] D2. Based on the scene feature information corresponding to each of the K first image groups, extract the description information of the first image in the K first image groups to obtain the scene feature description information corresponding to each of the K first image groups.
[0082] Based on the scene feature information corresponding to each of the K first image groups, obtain the description information corresponding to each scene feature information, so as to obtain the scene feature information corresponding to each of the K first image groups.
[0083] For example, to obtain the description information corresponding to the three elements "table", "chair", and "television", the description information for "table" is "a white square table placed in the middle of the north wall", the description information for "chair" is "a yellow round chair placed 5 meters south of the white square table", and the description information for "television" is "a black rectangular television placed in the center of the white square table". Then the scene feature description information for "there is a table, a chair and a television in the living room" is "a white square table placed in the middle of the north wall, a yellow round chair placed 5 meters south of the white square table, and a black rectangular television placed in the center of the white square table".
[0084] In this example, element extraction is performed on the scene information corresponding to each of the K first image groups to obtain scene feature information corresponding to each of the K first image groups. Description information of the scene feature information corresponding to each of the K first image groups is then obtained to obtain scene feature description information corresponding to each of the K first image groups. This improves the accuracy of the user status evaluation text obtained subsequently based on the scene feature description information corresponding to each of the K first image groups, thereby improving the accuracy of user status information acquisition.
[0085] In one possible implementation, K sub-user status evaluation texts are determined using scene feature information corresponding to each of the K first image groups. Each pair of sub-user status evaluation texts is then obtained to form a first association information set. The K sub-user status evaluation texts are then fused based on this first association information set to obtain the final user status evaluation text. The specific steps are as follows:
[0086] E1. Based on the scene feature description information corresponding to each of the K first image groups, determine K sub-user status evaluation texts, and the K sub-user status evaluation texts correspond one-to-one with the K first image groups.
[0087] K sub-user status evaluation texts can be determined based on the scene feature description information corresponding to each of the K first image groups. For example, if the scene feature description information is: "A white square table is placed in the middle position against the north wall", the sub-user status evaluation text determined based on this scene feature description information is: "What color is the table in the middle position against the north wall?"
[0088] E2. Obtain the association information between every two sub-user status evaluation texts in the K sub-user status evaluation texts to obtain the first association information set.
[0089] It can be that a loss value is calculated for every two sub-user status evaluation texts in K sub-user status evaluation texts. If the loss value of the two sub-user status evaluation texts is greater than or equal to a preset loss value threshold, the two sub-user status evaluation texts are considered to be related. Otherwise, the two sub-user status evaluation texts are not related, so as to obtain the first set of related information.
[0090] E3. Perform text fusion processing on the K sub-user status evaluation texts according to the first set of associated information to obtain the user status evaluation text.
[0091] The two corresponding sub-user status evaluation texts of all related information in the first association information set are fused to obtain the user status evaluation text corresponding to the target user.
[0092] In this example, K sub-user status assessment texts are determined based on the scene feature description information corresponding to each of the K first image groups. The correlation between each pair of sub-user status assessment texts is obtained to obtain a first correlation information set. Based on the first correlation information set, the related sub-user status assessment texts are fused to obtain the user status assessment text corresponding to the target user. This improves the accuracy of the obtained user status assessment texts and the accuracy of user status information acquisition.
[0093] In one possible implementation, the system obtains the historical user status information corresponding to the target user and the offset between this historical user status information and the current user status information. Based on this offset, the user status evaluation text is adjusted to obtain the target user's status adjustment task information. This adjusted status adjustment task information is then displayed to the target user. The specific steps are as follows:
[0094] F1. Obtain the historical user status information of the target user.
[0095] This could involve retrieving the target user's historical user status information from a storage device.
[0096] F2. Obtain the offset information between the historical user status information and the target user's user status information.
[0097] The offset information can be the difference between the target user's state information and historical user state information. This difference can be represented by the offset degree; for example, the larger the difference, the larger the offset, and vice versa.
[0098] F3. Determine the target user's status adjustment task information based on the offset information.
[0099] The user state assessment text for the target user can be adjusted based on the offset information to obtain the target user's state adjustment task information. If the deviation is negative, it indicates that the target user's state information has improved, and the training task can be reduced; if the deviation is positive, it indicates that the target user's state information has deteriorated, and the training task can be increased.
[0100] F4. Display the status adjustment task information.
[0101] This could involve displaying status adjustment task information through the client, for example, by playing it via voice.
[0102] In this example, by obtaining the target user's historical user status information and calculating the offset between the target user's historical user status information and the target user's current user status information, the offset information between the historical user status information and the target user's current user status information is obtained. Based on the offset information, the user status evaluation text is adjusted to obtain the target user's corresponding status adjustment task information, thereby improving the accuracy of the status adjustment task information.
[0103] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0104] In one embodiment, a user status information acquisition device is provided, which corresponds one-to-one with the user status information acquisition method in the above embodiments. For example... Figure 3 As shown, the user status information acquisition device includes a first acquisition unit 301, a classification unit 302, a second acquisition unit 303, a first determination unit 304, a second determination unit 305, and a third acquisition unit 306. Detailed descriptions of each functional module are as follows:
[0105] The first acquisition unit 301 is used to acquire the first memory data of the target user. The first memory data includes a first image set, and each first image in the first image set includes the target user.
[0106] The classification unit 302 is used to perform scene classification processing on the first image set to obtain K first image groups, and the scene information corresponding to each of the K first image groups is different;
[0107] The second acquisition unit 303 is used to acquire scene information corresponding to each of the K first image groups to obtain a first scene information set;
[0108] The first determining unit 304 is used to determine scene feature information corresponding to each of the K first image groups based on the first scene information set and the first images in the K first image groups.
[0109] The second determining unit 305 determines the user status evaluation text of the target user based on the scene feature information corresponding to each of the K first image groups.
[0110] The third acquisition unit 306 is used to acquire the user status information of the target user based on the user status evaluation text.
[0111] In one possible implementation, the third acquisition unit 306 is used for:
[0112] Display the user status assessment text;
[0113] Receive user status description information from the target user in response to the user status assessment text input;
[0114] Based on the user status description information, determine the user status information of the target user.
[0115] In one possible implementation, regarding the determination of the target user's user status information based on the user status description information, the third acquisition unit 306 is configured to:
[0116] The user status description information is segmented into sentences to obtain M first sentences;
[0117] Obtain the first semantic information corresponding to each of the M first statements;
[0118] Based on the first semantic information corresponding to the M first statements, the first reference user state information of the target user is determined;
[0119] Extract keywords from each of the M first statements to obtain the first keywords corresponding to each of the M first statements. Determine the first keywords corresponding to each of the M first statements as elements in the first keyword set to obtain the first keyword set.
[0120] Based on the set of first keywords corresponding to the M first statements, determine the second reference user status information of the target user;
[0121] The first reference user status information and the second reference user status information are fused together to obtain the user status information of the target user.
[0122] In one possible implementation, the third acquisition unit 306 is used to: fuse the first reference user state information and the second reference user state information to obtain the user state information of the target user.
[0123] The first reference user state information is encoded to obtain the first reference user state information vector;
[0124] The second reference user state information is encoded to obtain a second reference user state information vector;
[0125] The first reference user state information vector and the second reference user state information vector are fused to obtain the target user state information vector.
[0126] The user status information of the target user is determined based on the target user status information vector.
[0127] In one possible implementation, the first determining unit 304 is used for:
[0128] Based on the first scene information set, determine the scene feature information corresponding to each of the K first image groups;
[0129] Based on the scene feature information corresponding to each of the K first image groups, description information is extracted from the first image in the K first image groups to obtain scene feature description information corresponding to each of the K first image groups.
[0130] In one possible implementation, the second determining unit 305 is used for:
[0131] Based on the scene feature information corresponding to each of the K first image groups, K sub-user status evaluation texts are determined, and the K sub-user status evaluation texts correspond one-to-one with the K first image groups.
[0132] Obtain the association information between every two sub-user status evaluation texts in the K sub-user status evaluation texts to obtain a first association information set;
[0133] Based on the first set of associated information, the K sub-user status evaluation texts are fused to obtain the user status evaluation text.
[0134] In one possible implementation, after obtaining the target user's user status information based on the user status assessment text, the device is further configured to:
[0135] Obtain the historical user status information of the target user;
[0136] Obtain the offset information between the historical user status information and the target user's user status information;
[0137] The target user's status adjustment task information is determined based on the offset information;
[0138] Display the status adjustment task information.
[0139] In this example, the first memory data of the target user can be obtained. This first memory data includes a first image set, where each first image in the first image set includes the target user. Scene classification processing is performed on the first image set to obtain K first image groups. Each of the K first image groups corresponds to different scene information. Scene information corresponding to each of the K first image groups is obtained to form a first scene information set. Based on the first scene information set and the first images in the K first image groups, scene feature information corresponding to each of the K first image groups is determined. Based on each of the K first image groups… Based on the scene feature information corresponding to each of the first image groups, the user status assessment text of the target user is determined. The user status information of the target user is then obtained based on the user status assessment text. In this invention, the scene corresponding to the target user's first memory data can be classified to obtain K first image groups. Scene feature information is extracted from the K first image groups to obtain scene feature description information corresponding to each of the K first image groups. Based on the scene feature description information corresponding to each of the K first image groups, a user status assessment text corresponding to the target user is formulated. Therefore, the user status information of the target user can be accurately obtained by using the status assessment text, thus improving the accuracy of obtaining user status information.
[0140] Specific limitations regarding the user status information acquisition device can be found in the limitations of the user status information acquisition method described above, and will not be repeated here. Each module in the aforementioned user status information acquisition device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in the computer device, or stored in software in the memory of the computer device, so that the processor can call and execute the operations corresponding to each module.
[0141] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When executed by the processor, the computer program implements the functions or steps of a user status information acquisition method on the server side.
[0142] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:
[0143] Acquire the first memory data of the target user, the first memory data including a first image set, and the target user included in each first image in the first image set;
[0144] The first image set is subjected to scene classification processing to obtain K first image groups, and the scene information corresponding to each of the K first image groups is different;
[0145] Obtain scene information corresponding to each of the K first image groups to obtain a first scene information set;
[0146] Based on the first scene information set and the first images in the K first image groups, determine the scene feature information corresponding to each of the K first image groups;
[0147] Based on the scene feature information corresponding to each of the K first image groups, the user status evaluation text of the target user is determined;
[0148] The user status information of the target user is obtained based on the user status assessment text.
[0149] In this example, the first memory data of the target user can be obtained. This first memory data includes a first image set, where each first image in the first image set includes the target user. Scene classification processing is performed on the first image set to obtain K first image groups. Each of the K first image groups corresponds to different scene information. Scene information corresponding to each of the K first image groups is obtained to form a first scene information set. Based on the first scene information set and the first images in the K first image groups, scene feature information corresponding to each of the K first image groups is determined. Based on each of the K first image groups… Based on the scene feature information corresponding to each of the first image groups, the user status assessment text of the target user is determined. The user status information of the target user is then obtained based on the user status assessment text. In this invention, the scene corresponding to the target user's first memory data can be classified to obtain K first image groups. Scene feature information is extracted from the K first image groups to obtain scene feature description information corresponding to each of the K first image groups. Based on the scene feature description information corresponding to each of the K first image groups, a user status assessment text corresponding to the target user is formulated. Therefore, the user status information of the target user can be accurately obtained by using the status assessment text, thus improving the accuracy of obtaining user status information.
[0150] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:
[0151] Acquire the first memory data of the target user, the first memory data including a first image set, and the target user included in each first image in the first image set;
[0152] The first image set is subjected to scene classification processing to obtain K first image groups, and the scene information corresponding to each of the K first image groups is different;
[0153] Obtain scene information corresponding to each of the K first image groups to obtain a first scene information set;
[0154] Based on the first scene information set and the first images in the K first image groups, determine the scene feature information corresponding to each of the K first image groups;
[0155] Based on the scene feature information corresponding to each of the K first image groups, the user status evaluation text of the target user is determined;
[0156] The user status information of the target user is obtained based on the user status assessment text.
[0157] In this example, the first memory data of the target user can be obtained. This first memory data includes a first image set, where each first image in the first image set includes the target user. Scene classification processing is performed on the first image set to obtain K first image groups. Each of the K first image groups corresponds to different scene information. Scene information corresponding to each of the K first image groups is obtained to form a first scene information set. Based on the first scene information set and the first images in the K first image groups, scene feature information corresponding to each of the K first image groups is determined. Based on each of the K first image groups… Based on the scene feature information corresponding to each of the first image groups, the user status assessment text of the target user is determined. The user status information of the target user is then obtained based on the user status assessment text. In this invention, the scene corresponding to the target user's first memory data can be classified to obtain K first image groups. Scene feature information is extracted from the K first image groups to obtain scene feature description information corresponding to each of the K first image groups. Based on the scene feature description information corresponding to each of the K first image groups, a user status assessment text corresponding to the target user is formulated. Therefore, the user status information of the target user can be accurately obtained by using the status assessment text, thus improving the accuracy of obtaining user status information.
[0158] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server and client sides in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.
[0159] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0160] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0161] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A method for obtaining user status information, characterized in that, The method includes: Acquire the first memory data of the target user, the first memory data including a first image set, and the target user included in each first image in the first image set; The first image set is subjected to scene classification processing to obtain K first image groups, and the scene information corresponding to each of the K first image groups is different; Obtain scene information corresponding to each of the K first image groups to obtain a first scene information set; Based on the first scene information set and the first images in the K first image groups, determine the scene feature information corresponding to each of the K first image groups; Based on the scene feature information corresponding to each of the K first image groups, K sub-user status evaluation texts are determined, and the K sub-user status evaluation texts correspond one-to-one with the K first image groups. Obtain the association information between every two sub-user status evaluation texts in the K sub-user status evaluation texts to obtain a first association information set; Based on the first set of associated information, the K sub-user status evaluation texts are subjected to text fusion processing to obtain the user status evaluation text; Display the user status assessment text; Receive user status description information from the target user in response to the user status assessment text input; The user status description information is segmented into sentences to obtain M first sentences; Obtain the first semantic information corresponding to each of the M first statements; Based on the first semantic information corresponding to the M first statements, the first reference user state information of the target user is determined; Extract keywords from each of the M first statements to obtain the first keywords corresponding to each of the M first statements. Determine the first keywords corresponding to each of the M first statements as elements in the first keyword set to obtain the first keyword set. Based on the set of first keywords corresponding to the M first statements, determine the second reference user status information of the target user; The first reference user status information and the second reference user status information are fused together to obtain the user status information of the target user.
2. The user status information acquisition method according to claim 1, characterized in that, The step of fusing the first reference user status information and the second reference user status information to obtain the user status information of the target user includes: The first reference user state information is encoded to obtain the first reference user state information vector; The second reference user state information is encoded to obtain a second reference user state information vector; The first reference user state information vector and the second reference user state information vector are fused to obtain the target user state information vector. The user status information of the target user is determined based on the target user status information vector.
3. The user status information acquisition method according to claim 1 or 2, characterized in that, The step of determining scene feature description information corresponding to each of the K first image groups based on the first scene information set and the first images in the K first image groups includes: Based on the first scene information set, determine the scene feature information corresponding to each of the K first image groups; Based on the scene feature information corresponding to each of the K first image groups, description information is extracted from the first image in the K first image groups to obtain scene feature description information corresponding to each of the K first image groups.
4. The user status information acquisition method according to claim 1, characterized in that, After obtaining the target user's user status information based on the user status assessment text, the method further includes: Obtain the historical user status information of the target user; Obtain the offset information between the historical user status information and the target user's user status information; The target user's status adjustment task information is determined based on the offset information; Display the status adjustment task information.
5. A user status information acquisition device, characterized in that, The user status information acquisition device includes: The first acquisition unit is used to acquire the first memory data of the target user, the first memory data including a first image set, and each first image in the first image set including the target user; A classification unit is used to perform scene classification processing on the first image set to obtain K first image groups, and the scene information corresponding to each of the K first image groups is different; The second acquisition unit is used to acquire scene information corresponding to each of the K first image groups to obtain a first scene information set; The first determining unit is configured to determine scene feature information corresponding to each of the K first image groups based on the first scene information set and the first images in the K first image groups. The second determining unit is configured to determine K sub-user status evaluation texts based on scene feature information corresponding to each of the K first image groups, wherein the K sub-user status evaluation texts correspond one-to-one with the K first image groups; obtain association information between every two sub-user status evaluation texts in the K sub-user status evaluation texts to obtain a first association information set; and perform text fusion processing on the K sub-user status evaluation texts based on the first association information set to obtain the user status evaluation text. The third acquisition unit is used to display the user status assessment text; receive user status description information input by the target user in response to the user status assessment text; perform sentence segmentation on the user status description information to obtain M first statements; acquire first semantic information corresponding to each of the M first statements; determine the first reference user status information of the target user based on the first semantic information corresponding to each of the M first statements; extract keywords from each of the M first statements to obtain first keywords corresponding to each of the M first statements, and determine the first keywords corresponding to each of the M first statements as elements in a first keyword set to obtain a first keyword set; determine the second reference user status information of the target user based on the first keyword set corresponding to the M first statements; and perform fusion processing on the first reference user status information and the second reference user status information to obtain the user status information of the target user.
6. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the user status information acquisition method as described in any one of claims 1 to 4.
7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the user status information acquisition method as described in any one of claims 1 to 4.