Information reminding method and device and electronic equipment

Abstract

The invention discloses an information reminding method and device and electronic equipment, and relates to the technical field of electronic information. The method comprises the following steps: determining at least one piece of notification information obtained by a user terminal at a target moment; obtaining a user behavior spectrogram within a specific duration before the target moment, wherein the user behavior spectrogram is used for representing behavior activities of a user for at least one application program in the user terminal within the specific duration; and determining target reminding information for reminding the user based on the user behavior spectrogram and the at least one piece of notification information.

Description

Technical Field

[0001] The present application relates to the field of electronic information technology, and in particular to an information reminder method, device and electronic equipment. Background Art

[0002] Mobile assistant is a must-have application for today's smartphones. In addition to helping users control devices, query information and chat, it also plays an important role in providing smart reminders, such as timely reminders of users' travel information and schedule information.

[0003] In the related art, the reminder information is usually obtained by binding the service account of the relevant mobile application (Application, APP) in the mobile assistant, or by parsing the interface information of the relevant APP to generate the reminder card of the mobile assistant. Although these two methods realize the corresponding reminder function, they are both limited by the APP of the information source, and some non-APP reservation information cannot be obtained, such as booking a flight or hotel by phone or email. At this time, the corresponding reservation information cannot be obtained through the relevant APP, so the relevant intelligent reminder cannot be given to the user, resulting in poor user experience and low universality. Summary of the invention

[0004] In order to solve the above technical problems, the embodiments of the present application provide an information reminder method, device and electronic device.

[0005] The technical solution of the embodiment of the present application is implemented as follows:

[0006] In a first aspect, an embodiment of the present application provides an information reminder method, comprising:

[0007] Determine at least one notification information obtained by the user terminal at a target time;

[0008] Obtaining a user behavior spectrum within a specific time period before a target moment, where the user behavior spectrum is used to characterize the user's behavior activities for at least one application in the user terminal within the specific time period;

[0009] Based on the user behavior spectrum and at least one notification information, target reminder information for reminding the user is determined.

[0010] In some embodiments, the specific duration includes at least one time slice; and the information reminder method further includes:

[0011] Determine user behavior information corresponding to each time slice, where the user behavior information is used to represent the user's behavior activities for at least one application in the user terminal in each time slice;

[0012] Based on the user behavior information corresponding to each time slice, a user behavior spectrum is determined.

[0013] In some embodiments, based on the user behavior information corresponding to each time slice, determining the user behavior spectrum includes:

[0014] Determine a first vector corresponding to the user behavior information corresponding to each time slice;

[0015] A user behavior spectrum is determined based on the first vectors corresponding to each time slice.

[0016] In some embodiments, the information reminder method further includes:

[0017] Determine the target application triggered by the user in each time slice based on the user behavior information corresponding to each time slice;

[0018] Determine text introduction information of the target application triggered by the user in each time slice;

[0019] Determine a second vector corresponding to the text introduction information of the target application corresponding to each time slice;

[0020] The determining the user behavior spectrum based on the first vectors respectively corresponding to the time slices includes:

[0021] A user behavior spectrum is determined based on the first vector and the second vector corresponding to each time slice.

[0022] In some embodiments, for different target applications having the same function, the second vectors corresponding to the text introduction information of the different target applications have similarities.

[0023] In some embodiments, based on the user behavior spectrum and at least one notification information, determining target reminder information for reminding the user includes:

[0024] Determine a first target intention of the user based on the user behavior spectrum and at least one notification information;

[0025] Based on the first target intention, target reminder information is determined.

[0026] In some embodiments, determining the first target intention of the user based on the user behavior spectrum and at least one notification information includes:

[0027] Determining at least one second target intention of the user based on the user behavior spectrum;

[0028] Based on the at least one second target intent and the at least one notification information, a first target intent of the user is determined.

[0029] In some embodiments, determining the first target intent of the user based on the at least one second target intent and the at least one notification information includes:

[0030] Based on at least one second target intention, determining a confidence level corresponding to at least one notification information;

[0031] Based on the confidence level respectively corresponding to at least one notification information, the user's first target intention is determined.

[0032] In a second aspect, an embodiment of the present application provides an information reminder device, including:

[0033] A first determining module, used to determine at least one notification information obtained by the user terminal at a target time;

[0034] An acquisition module, used to obtain a user behavior spectrum within a specific time period before a target moment, where the user behavior spectrum is used to characterize the user's behavior activities for at least one application in the user terminal within the specific time period;

[0035] The second determination module is used to determine target reminder information for reminding the user based on the user behavior spectrum and at least one notification information.

[0036] In a third aspect, an embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory is used to store executable data instructions; when the processor is used to execute the executable data instructions stored in the memory, the information reminder method provided in the first aspect is implemented.

[0037] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the information reminder method provided in the first aspect is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, a brief introduction will be given below to the drawings required for use in the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 A schematic diagram of a mobile phone assistant reminder function provided for related technologies;

[0040] Figure 2 A schematic diagram of a related technology for realizing information reminder by binding the service account of a related application in a mobile assistant;

[0041] Figure 3 One of the flow charts of an information reminder method provided in an embodiment of the present application;

[0042] Figure 4 A second flowchart of an information reminder method provided in an embodiment of the present application;

[0043] Figure 5 A third flow chart of an information reminder method provided in an embodiment of the present application;

[0044] Figure 6 A fourth flowchart of an information reminder method provided in an embodiment of the present application;

[0045] Figure 7 A schematic diagram of a structural flow of an information reminder provided in an embodiment of the present application;

[0046] Figure 8 A schematic diagram of the structure of an information reminder device provided in an embodiment of the present application;

[0047] Fig. 9 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present application.

[0049] It should be noted that in the description of the embodiments of the present application, the terms "first", "second", etc. are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more.

[0050] In order to facilitate a clearer understanding of the embodiments of the present application, some relevant background knowledge is first introduced as follows.

[0051] The Large Language Model (LLM), based on the Generative Artificial Intelligence (Generative AI) framework, gives mobile assistants powerful understanding, generation and interaction capabilities. Mobile assistants are used to provide timely reminders of users' travel information and schedule information, and most of this information is closely related to users' personal privacy and is not suitable for cloud-based solutions. As a mobile phone manufacturer, mobile assistants can be deeply integrated with the underlying hardware of the mobile phone, and obtain notification information from different sources including text messages, and can also obtain user usage behavior. It is very important for mobile phone products to use the information on the device side to realize AI functions without being connected to the Internet.

[0052] In the related technology, users are usually required to set the ticket purchase information or reservation information by themselves. The mobile assistant can parse the information set by the user and then remind them, such as iPhone reminders, Huawei Jingye signature and moto mobile travel assistant. This reminder solution requires manual input by the user and is not very intelligent. Or follow the example Figure 1 or Figure 2 The mobile assistant reminder method shown in the figure realizes information reminder. Figure 1 As shown in the figure, users can set up a booking assistant (i.e., mobile assistant) application. After the user completes the corresponding ticket purchase or reservation using a certain application, the booking assistant can automatically generate an assistant card to remind the user by parsing the content of the application information interface. This reminder solution is affected by the changes in the application information interface and can only support a limited number of applications. Some non-application reservations cannot be supported, such as phone, email, and WeChat. Figure 2 As shown in the figure, users can obtain reminder information by binding the service account of related applications in the mobile assistant, such as OPPO's Xiaobu Assistant. This reminder solution is limited by the application and is not a purely offline end-side solution.

[0053] In order to overcome at least some of the above-mentioned defects existing in the related art, the embodiments of the present application provide an information reminder method, device and electronic device, which can accurately determine the target reminder information used to remind the user based on at least one notification information obtained by the user terminal at the target time and a user behavior spectrum that can characterize the user's behavioral activities for at least one application in the user terminal within a specific time period, and is not limited by the application source of the notification information, and has stronger universality.

[0054] The following is an illustrative introduction to the information reminder method, device and electronic device provided in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application.

[0055] Figure 3One of the flow charts of an information reminder method provided in an embodiment of the present application is as follows: Figure 3 As shown, the method includes:

[0056] S301. Determine at least one notification information obtained by a user terminal at a target time.

[0057] It should be noted that the user terminal may include various types of devices that have the function of obtaining or receiving notification information, such as a smart phone, a tablet computer, a smart watch, a laptop computer, a desktop computer, a smart TV, or a smart speaker.

[0058] It should be noted that the notification information obtained by the user terminal at the target time may include multiple types, and these notification information are intended to convey important or interesting content to the user. For example, the notification information may include at least:

[0059] System notifications: including system update notifications, system status notifications (e.g. low battery, insufficient storage space) and system security notifications (e.g. virus protection, account security), etc.

[0060] Message notifications: including SMS notifications, instant messaging notifications (such as message reminders from instant messaging applications such as WeChat and QQ), and email notifications;

[0061] App notifications: including in-app message notifications (e.g., news push notifications in news apps, discount information in shopping apps, schedule reminders in calendar apps, etc.) and push notifications (notifications sent by apps through the system push mechanism, usually appearing in the device's notification bar or lock screen), etc.

[0062] Service notifications: including order notifications (such as order status updates on e-commerce platforms, logistics information notifications, booked flights, trains, movie performances, hotels, restaurants, hospital registrations and living expenses, etc.), payment notifications (such as payment success, payment failure and refund notifications, etc.), service reminders (such as appointment service reminders, membership expiration reminders and birthday reminders, etc.).

[0063] It should be noted that the target time can be any time, and the embodiments of the present application do not specifically limit this.

[0064] In some embodiments, all or part of the notification information obtained by the user terminal at the target time may be determined.

[0065] S302: Obtain a user behavior spectrum within a specific time period before the target moment, where the user behavior spectrum is used to characterize the user's behavior activities with respect to at least one application in the user terminal within the specific time period.

[0066] It should be noted that the user behavior spectrum represents the user's behavioral activities on at least one application in the user terminal within a specific period of time in a content-based manner, including information such as page / application dwell time, text box input, screen clicks, scrolling and sliding pages, and is not limited to traditional usage time, number of words and times.

[0067] It should be noted that the specific time before the target time can be adaptively set based on actual applications, and the present embodiment does not specifically limit this, for example, 20 minutes before time a, 1 hour before time b, 5 hours before time c, etc.

[0068] In some embodiments, the horizontal axis of the user behavior spectrogram can represent time, and the vertical axis can represent the embedding vector (Embedding) of the user behavior activity within a specific time period. It can be understood that the selection of the horizontal axis scale reflects the time resolution, and the granularity of the vertical axis application reflects the spatial (application usage) resolution. In this way, the user behavior spectrogram is a non-sparse matrix with an adaptive time scale, which is a compressed representation of the user's historical behavior activities.

[0069] For example, first, determine at least one notification information obtained by the user terminal at time a, and then obtain a user behavior spectrum within 20 minutes before time a, which characterizes the user's behavioral activities such as page dwelling, text box input, and screen clicks for at least one application in the user terminal within 20 minutes before time a.

[0070] S303: Determine target reminder information for reminding the user based on the user behavior spectrum and the at least one notification information.

[0071] In some embodiments, based on the user behavior spectrum, the target notification information that needs to remind the user can be determined from at least one notification information, and then key information (such as the train number information, seat information and travel time information of the user's train ticket booking, etc.) can be extracted from each target notification information as target reminder information.

[0072] In some embodiments, after it is determined that the target reminder information is obtained, the target reminder information can be displayed in a pop-up window or a floating window to achieve the purpose of reminding the user.

[0073] In some embodiments, after it is determined that the target reminder information is obtained, the target reminder information can be broadcasted by voice broadcast to achieve the purpose of reminding the user.

[0074] It should be noted that the information reminder method provided in the embodiment of the present application is a purely offline terminal-side information reminder method, that is, the target reminder information used to remind the user is determined by utilizing the notification information obtained by the user terminal and the user behavior spectrum that can characterize the user's behavior activities. This method can effectively remind the user without a network connection.

[0075] The information reminder method provided in the embodiment of the present application first determines at least one notification information obtained by the user terminal at the target time, and obtains a user behavior spectrum that can characterize the user's behavioral activities with respect to at least one application in the user terminal within a specific period of time before the target time, and then based on the user behavior spectrum and at least one notification information obtained by the user terminal, it can accurately determine the target reminder information for reminding the user, so as to provide the user with relevant intelligent reminders, and is not limited by the application source of the notification information, has stronger universality, and provides a better user experience.

[0076] In some embodiments, the specific duration includes at least one time slice;

[0077] The information reminder method also includes:

[0078] Determine user behavior information corresponding to each of the time slices, where the user behavior information is used to characterize the behavior activities of the user with respect to at least one application in the user terminal in each of the time slices;

[0079] The user behavior spectrum is determined based on the user behavior information corresponding to each of the time slices.

[0080] In an embodiment of the present application, the specific duration before the target moment can be divided into at least one time slice, and then the user behavior information corresponding to each time slice is determined, and then based on the user behavior information corresponding to each time slice, a user behavior spectrum is determined, wherein the user behavior information corresponding to each time slice can characterize the user's behavioral activities for at least one application in the user terminal in each time slice.

[0081] For example, the hour before time b can be divided into three time slices, each time slice is 20 minutes, and then the user behavior information corresponding to each of the three time slices is determined, and then the user behavior information corresponding to each time slice is converted into an embedding vector, and then a user behavior spectrum is constructed with the three time slices as the horizontal axis and the embedding vectors of the user behavior information corresponding to each time slice as the vertical axis.

[0082] It should be noted that the duration of the time slice division can be adaptively set based on the actual application. The embodiment of the present application does not make any specific limitations on this. Finer or coarser time granularity can be divided according to specific needs.

[0083] It should be noted that the user behavior information corresponding to each time slice may include the applications opened by the user, the operations in each application (such as clicking, scrolling, and inputting, etc.), and the application usage time.

[0084] It can be understood that the embodiment of the present application divides the specific time before the target moment into multiple time slices, and based on the user behavior information corresponding to each time slice, the user's behavioral activities in different time periods can be determined more finely, which helps to capture subtle differences in user behavior, and thus enables the obtained user behavior spectrum to more accurately characterize the user's behavioral activities for at least one application in the user terminal within the specific time before the target moment.

[0085] In some embodiments, determining the user behavior spectrum based on the user behavior information corresponding to each of the time slices includes:

[0086] Determine a first vector corresponding to the user behavior information corresponding to each of the time slices;

[0087] The user behavior spectrogram is determined based on the first vectors respectively corresponding to the time slices.

[0088] In an embodiment of the present application, after obtaining the user behavior information corresponding to each time slice, the user behavior information corresponding to each time slice can be converted into a vector form, that is, a first vector, and then a user behavior spectrum graph is constructed using the first vectors corresponding to each time slice.

[0089] It should be noted that the first vector corresponding to the user behavior information may be multi-dimensional, with each dimension corresponding to a behavior feature. The value of the first vector may be a specific numerical value (such as usage time, number of clicks, etc.), or a value after normalization or standardization.

[0090] In some embodiments, the first vectors corresponding to each time slice can be combined to form a two-dimensional matrix or a higher-dimensional data structure. Each row of this data structure represents the first vector of a time slice, and each column represents a specific behavior feature. Furthermore, based on this data structure, a user behavior spectrum can be constructed.

[0091] In some embodiments, the user behavior spectrogram can be in various forms, including but not limited to the following: time series graph: use time as the horizontal axis and one or some behavioral features as the vertical axis to draw a trend graph of these features changing over time; heat map: use color depth to represent the intensity or frequency of behavioral features in different time slices, so as to intuitively show the changes in user behavior activities; radar chart: for each time slice, use multiple behavioral features as multiple dimensions of the radar chart, and form a polygon by connecting the points on these dimensions to show the user's behavior activities in the time slice; scatter plot or cluster plot: if there are many time slices, you can use a scatter plot or cluster plot to show the user's behavior activities in different time slices.

[0092] It can be understood that the embodiment of the present application determines a user behavior spectrum by first vectors corresponding to user behavior information corresponding to each time slice. Through the user behavior spectrum, the user's behavior activities in different time slices can be intuitively determined.

[0093] In some embodiments, the information reminder method further includes:

[0094] Determining a target application triggered by the user in each of the time slices based on the user behavior information corresponding to each of the time slices;

[0095] Determining text introduction information of the target application triggered by the user in each time slice;

[0096] Determine a second vector corresponding to the text introduction information of the target application corresponding to each of the time slices;

[0097] The determining the user behavior spectrum based on the first vectors respectively corresponding to the time slices includes:

[0098] The user behavior spectrogram is determined based on the first vector and the second vector respectively corresponding to each of the time slices.

[0099] In an embodiment of the present application, after obtaining the user behavior information corresponding to each time slice, the target application triggered by the user in each time slice can be determined based on the user behavior information corresponding to each time slice, and then the text introduction information of the target application triggered by the user in each time slice can be determined, and then the text introduction information of the target application triggered by the user in each time slice can be converted into a second vector, and then based on the first vector and the second vector corresponding to each time slice, a user behavior spectrum can be determined.

[0100] It should be noted that since the user behavior information corresponding to each time slice includes the applications opened by the user, the operations in each application (such as clicking, scrolling and inputting, etc.) and the application usage time, the target application triggered by the user in each time slice can be determined based on the user behavior information corresponding to each time slice.

[0101] It should be noted that the text introduction information of the target application triggered by the user in each time slice can come from the metadata of each application (such as the description in the application store), the help document in the application, or the user-defined label.

[0102] In some embodiments, the text introduction information of the target application can be cleaned to remove useless information (such as advertising words, special characters, etc.), and standardized (such as unifying the text format, converting to lowercase, etc.), and then using natural language processing technology, the cleaned and standardized text introduction information can be converted into a vector form, i.e., a second vector, which can capture the semantic information in the text introduction information.

[0103] In some embodiments, first, the first vector and the second vector corresponding to each time slice can be associated to form a complete behavior representation of each time slice; second, the associated vectors are mapped to the time slice to ensure that each time slice has a unique vector representation; then, the unique vector representations corresponding to each time slice are combined to form a two-dimensional matrix or a higher-dimensional data structure, in which each row represents a unique vector representation of a time slice and each column represents a specific behavior feature. Furthermore, based on this data structure, a user behavior spectrum can be constructed.

[0104] In some embodiments, the first vector and the second vector may be associated by concatenation, weighted summation, or other methods, which is not specifically limited in the embodiments of the present application.

[0105] It can be understood that the embodiment of the present application, by combining the first vector corresponding to the user behavior information and the second vector corresponding to the text introduction information of the target application, can make the obtained user behavior spectrum more comprehensive and accurate in reflecting the user's behavioral activities in different time slices as well as the user's interests, needs and preferences.

[0106] In some embodiments, for different target applications having the same function, the second vectors corresponding to the text introduction information of the different target applications have similarities.

[0107] It should be noted that different applications may have the same functions. For example, WeChat and Alipay both have the function of payment; for another example, Tencent Video and iQiyi Video both have the function of watching videos. When different applications have the same or similar functions, their text introduction information often contains some common keywords, phrases or descriptions. These common elements will be mapped to similar vector spaces during the text vectorization process, so that the corresponding second vectors have similarity.

[0108] For example, Tencent Video and iQiyi Video, as video viewing platforms, both contain keywords related to video viewing in their text descriptions, such as "HD video", "online viewing", "movies" and "TV series", etc. The similarity of these keywords will be reflected in the corresponding second vector.

[0109] In some embodiments, the user behavior spectrum may be determined based on the following steps, including:

[0110] (1) For a time slot, the user behavior information corresponding to the time slot is expressed in text form, such as “click / Ctrip Travel / #3 click / air ticket / #2 input / Beijing Shanghai / #1 click / query / #2 stay / 45 seconds / #2 exit / #3 click / 12306 / #3 input / Shanghai Beijing / #1 click / query ticket / #2 click / book / #2 click / submit order / #2…”;

[0111] (2) Convert the user behavior information represented in text form in (1) into an M-dimensional embedding vector;

[0112] (3) Convert the textual introduction information corresponding to the application used in (1) into an M-dimensional embedding vector in the same way as (2);

[0113] It should be noted that this step is introduced to take into account the different preferences of different users in using applications. For example, some users are accustomed to using WeChat payment, while some users are accustomed to using Alipay payment. Since both WeChat and Alipay have payment functions, the vector converted from the text introduction information of WeChat is similar to the vector converted from the text introduction information of Alipay.

[0114] (4) Weightedly merge the embedding vectors obtained in (2) and (3) to obtain an M-dimensional vector, i.e., a high-dimensional representation of the content of the timeslot behavior;

[0115] (5) N time slices are used to obtain a two-dimensional array of N×M dimensions, which represents the user behavior spectrum with time series. Where N and M are both positive integers.

[0116] It can be understood that the embodiment of the present application determines the user behavior spectrum by combining the first vector corresponding to the user behavior information and the second vector corresponding to the text introduction information of the target application, and the second vectors corresponding to the text introduction information of different target applications with the same function have similarities. Not only can a user behavior spectrum that can reflect the user portrait be constructed from the two levels of behavior and semantics, but this multi-dimensional portrait can also more comprehensively reflect the user's interests and preferences.

[0117] In some embodiments, determining target reminder information for reminding the user based on the user behavior spectrum and the at least one notification information includes:

[0118] Determining a first target intention of the user based on the user behavior spectrum and the at least one notification information;

[0119] Based on the first target intention, the target reminder information is determined.

[0120] In an embodiment of the present application, the user behavior spectrum can be analyzed to determine the user's historical behavior activities and preferences, and then the notification information matching the user's historical behavior activities and preferences can be extracted from at least one notification information obtained by the user terminal at the target time to determine the user's first target intention, such as checking flight information, checking hotel reservation information, etc., and then based on the first target intention, the target reminder information for reminding the user is determined.

[0121] For example, Figure 4 A second flow chart of an information reminder method provided in an embodiment of the present application is as follows: Figure 4 As shown, the method includes:

[0122] S401, determining at least one notification information obtained by a user terminal at a target time;

[0123] S402, obtaining a user behavior spectrum within a specific time period before the target time, wherein the user behavior spectrum is used to characterize the user's behavior activities with respect to at least one application in the user terminal within the specific time period;

[0124] S403, determining the first target intention of the user based on the user behavior spectrum and the at least one notification information;

[0125] S404: Determine target reminder information for reminding the user based on the first target intention.

[0126] It should be noted that, for the description of the same steps and the same contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0127] It can be understood that the embodiments of the present application, by combining a user behavior spectrum that can characterize the user's historical behavior activities and at least one notification information obtained by the user terminal at the target time, can more accurately infer the user's real needs or intentions, that is, the user's first target intention, and then based on the user's first target intention, can accurately determine the target reminder information used to remind the user, thereby improving the accuracy of the reminder.

[0128] In some embodiments, determining the first target intention of the user based on the user behavior spectrum and the at least one notification information includes:

[0129] Based on the user behavior spectrum, determining at least one second target intention of the user;

[0130] Based on the at least one second target intent and the at least one notification information, a first target intent of the user is determined.

[0131] In an embodiment of the present application, the user behavior spectrum can be analyzed to determine the user's historical behavior activities and preferences, and then based on the user's historical behavior activities and preferences, at least one second target intention of the user can be determined, such as checking flight information, checking hotel reservation information, etc.; then, the user's first target intention can be determined by combining the user's at least one second target intention and at least one notification information obtained by the user terminal at the target time.

[0132] It can be understood that the first target intention may be part of or all of the at least one second target intention.

[0133] In some embodiments, the correlation between each notification information and each second target intention can be evaluated, wherein notification information with high correlation is more likely to trigger the user's first target intention.

[0134] For example, Figure 5 A third flow chart of an information reminder method provided in an embodiment of the present application is as follows: Figure 5 As shown, the method includes:

[0135] S501, determining at least one notification information obtained by a user terminal at a target time;

[0136] S502, obtaining a user behavior spectrum within a specific time period before the target time, wherein the user behavior spectrum is used to characterize the user's behavior activities with respect to at least one application in the user terminal within the specific time period;

[0137] S503, determining at least one second target intention of the user based on the user behavior spectrum;

[0138] S504. Determine the first target intent of the user based on the at least one second target intent and the at least one notification information.

[0139] S505: Based on the first target intention, determine target reminder information for reminding the user.

[0140] It should be noted that, for the description of the same steps and the same contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0141] It can be understood that the embodiment of the present application determines at least one second target intention of the user based on the user behavior spectrum, and then combines the at least one second target intention of the user and at least one notification information obtained by the user terminal at the target time, so as to deeply understand the user's potential needs, thereby accurately determining the user's first target intention.

[0142] In some embodiments, the determining the first target intent of the user based on the at least one second target intent and the at least one notification information includes:

[0143] Determining, based on the at least one second target intention, a confidence level corresponding to each of the at least one notification information;

[0144] Based on the confidence levels respectively corresponding to the at least one notification information, the first target intention of the user is determined.

[0145] In an embodiment of the present application, for each notification information obtained by the user terminal at the target time, the degree of association between it and each second target intention can be analyzed, and then based on the association analysis result, a corresponding confidence level can be calculated for each notification information, and then all notification information can be sorted according to their corresponding confidence levels. Notification information with high confidence levels is more likely to be related to the user's first target intention, thereby determining the user's first target intention.

[0146] It should be noted that confidence is a quantitative indicator that can be used to indicate the degree of match or credibility between each notification information and each second target intention. The confidence calculation method can be adaptively designed based on actual applications, and may involve weighted summation of multiple factors or prediction of machine learning models.

[0147] It should be noted that the degree of correlation between each notification information and each second target intention can be analyzed in a variety of ways, such as text similarity analysis, keyword matching and context understanding, etc., in order to evaluate whether the notification information is related to one or more second target intentions of the user, and the degree of correlation.

[0148] It can be understood that in the embodiments of the present application, based on the user behavior spectrum of the user in the recent specific time period, on the one hand, at least one second target intention of the user can be determined, and then based on the at least one second target intention, the confidence level can be calculated for notification information based on various sources such as text messages, local notifications, voice phone chats and emails, and on the other hand, information from different sources can be associated.

[0149] For example, Figure 6 A fourth flow chart of an information reminder method provided in an embodiment of the present application is as follows: Figure 6 As shown, the method includes:

[0150] S601, determining at least one notification information obtained by a user terminal at a target time;

[0151] S602: Obtain a user behavior spectrum within a specific time period before the target time, wherein the user behavior spectrum is used to characterize the user's behavior activities with respect to at least one application in the user terminal within the specific time period;

[0152] S603: Determine at least one second target intention of the user based on the user behavior spectrum;

[0153] S604: Determine, based on the at least one second target intention, a confidence level corresponding to each of the at least one notification information;

[0154] S605: Determine the first target intention of the user based on the confidence level corresponding to each of the at least one notification information;

[0155] S606: Based on the first target intention, determine target reminder information for reminding the user.

[0156] It should be noted that, for the description of the same steps and the same contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0157] It can be understood that the embodiment of the present application determines the confidence level corresponding to at least one notification message based on at least one second target intention. The confidence level can represent the degree of match or credibility between each notification message and each second target intention. Further, based on the confidence level corresponding to each notification message, the user's first target intention can be accurately determined.

[0158] For example, Figure 7 A structural flow diagram of an information reminder provided in an embodiment of the present application, such as Figure 7As shown, the structure includes an Encoder module (encoding module), an intention understanding module and an information extraction module. Among them, the Encoder module is used to encode the user behavior spectrum within a specific time before the target moment and predict the user intention; the intention understanding module is used to further predict the user intention based on the user intention predicted by the Encoder module and at least one notification information obtained by the user terminal at the target moment; the information extraction module is used to extract information based on the user intention predicted by the intention understanding module, and the application layer presents the extracted key information (target reminder information) in the form of cards. Among them, Figure 7 The long-term user behavior spectrogram in is a behavior spectrogram that can characterize the user's long-term behavior activities before the target moment. The horizontal axis of the behavior spectrogram represents N time slots, and the vertical axis represents the M-dimensional embedding vector of the user's behavior activities in each time slot. The long-term user behavior spectrogram is a high-level content representation of the user's historical behavior activities, which contains user intention information. For example, opening the Amap application to search for nearby theaters and opening the Maoyan Movie application to search for the latest movies can predict that the user has the intention to buy movie tickets. When the user terminal receives a notification message, the intention prediction of the user's historical behavior activities can well enhance the user's intention judgment. The user behavior spectrogram is input into the Encoder module for encoding. The Encoder module can predict multiple intentions corresponding to each time slot, and then find the best intention prediction path in the entire space, which can include multiple intentions, such as booking a hotel + booking a flight ticket, as part of the input of the intention understanding module to improve accuracy.

[0159] It is understandable that the embodiments of the present application can at least bring the following beneficial effects:

[0160] (1) It is not limited by the type of application and the source of notification information and has greater universality.

[0161] (2) In view of the requirement that intelligent reminders cannot have false alarms, the embodiment of the present application takes into account the user's long-term behavioral activities and proposes a user behavior spectrum to enhance the accuracy of user intention judgment, effectively reduce false alarms, and provide users with more accurate intelligent reminder services.

[0162] It should be noted that with the reduction of communication costs, today's mobile phone users receive a large number of notification messages (such as advertisements and warning messages, etc.) every day. These messages come in various forms, and some of them are easily confused with real effective reminder messages. Therefore, it is easy to generate false alarms and affect user experience by only using the content of notification messages to remind users. When users need to book air tickets and hotels for business trips, they will click on relevant applications to inquire before booking, and there will be payment operations after booking. The use of this information will make the acquisition of key reminder information more accurate and avoid interference from other highly similar non-effective notification messages.

[0163] (3) The information reminder method provided in the embodiment of the present application is an offline solution, which does not increase the complexity of the natural language understanding (NLU) model and is suitable for running on the device side.

[0164] The information reminder device provided in the embodiment of the present application is described below. The information reminder device described below and the information reminder method described above can refer to each other.

[0165] Figure 8 A schematic diagram of the structure of the information reminder device provided in the embodiment of the present application, such as Figure 8 As shown, the device includes: a first determination module 810, an acquisition module 820 and a second determination module 830; wherein:

[0166] A first determining module 810, configured to determine at least one notification information obtained by a user terminal at a target time;

[0167] An acquisition module 820, configured to acquire a user behavior spectrum within a specific time period before the target moment, wherein the user behavior spectrum is used to characterize the user's behavior activities with respect to at least one application in the user terminal within the specific time period;

[0168] The second determination module 830 is used to determine target reminder information for reminding the user based on the user behavior spectrum and the at least one notification information.

[0169] The information reminder device provided in the embodiment of the present application first determines at least one notification information obtained by the user terminal at the target time, and obtains a user behavior spectrum that can characterize the user's behavioral activities with respect to at least one application in the user terminal within a specific period of time before the target time, and then based on the user behavior spectrum and at least one notification information obtained by the user terminal, it can accurately determine the target reminder information for reminding the user, so as to provide the user with relevant intelligent reminders, and is not limited by the application source of the notification information, has stronger universality, and provides a better user experience.

[0170] In some embodiments, the specific duration includes at least one time slice;

[0171] The device also includes:

[0172] A third determining module is used to determine user behavior information corresponding to each of the time slices, where the user behavior information is used to characterize the behavior activities of the user for at least one application in the user terminal in each of the time slices;

[0173] The fourth determination module is used to determine the user behavior spectrum based on the user behavior information corresponding to each of the time slices.

[0174] In some embodiments, the fourth determining module includes:

[0175] A first determining unit, configured to determine a first vector corresponding to the user behavior information corresponding to each of the time slices;

[0176] The second determining unit is used to determine the user behavior spectrum based on the first vectors corresponding to each of the time slices.

[0177] In some embodiments, the apparatus further comprises:

[0178] A fifth determination module, configured to determine a target application triggered by the user in each of the time slices based on the user behavior information corresponding to each of the time slices;

[0179] A sixth determining module, used to determine the text introduction information of the target application triggered by the user in each time slice;

[0180] A seventh determination module, used to determine the second vector corresponding to the text introduction information of the target application corresponding to each of the time slices;

[0181] The second determination unit is further used to determine the user behavior spectrum based on the first vector and the second vector corresponding to each of the time slices.

[0182] In some embodiments, for different target applications having the same function, the second vectors corresponding to the text introduction information of the different target applications have similarities.

[0183] In some embodiments, the second determining module 830 includes:

[0184] A third determining unit, configured to determine a first target intention of the user based on the user behavior spectrum and the at least one notification information;

[0185] The fourth determining unit is used to determine the target reminder information based on the first target intention.

[0186] In some embodiments, the third determining unit is further configured to:

[0187] Based on the user behavior spectrum, at least one second target intention of the user is determined; based on the at least one second target intention and the at least one notification information, the first target intention of the user is determined.

[0188] In some embodiments, the third determining unit is further configured to:

[0189] Based on the at least one second target intention, the confidence level corresponding to each of the at least one notification information is determined; based on the confidence level corresponding to each of the at least one notification information, the first target intention of the user is determined.

[0190] It should be noted here that the above-mentioned information reminder device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned information reminder method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0191] Fig. 9 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application, such as Fig. 9 As shown, the electronic device may include: a processor 910, a communication interface 920, a memory 930 and a communication bus 940, wherein the processor 910, the communication interface 920 and the memory 930 communicate with each other through the communication bus 940. The processor 910 may call the executable data instructions stored in the memory 930 to execute some or all of the steps in the information reminder method provided in the above embodiments.

[0192] In addition, the executable data instructions stored in the above-mentioned memory 930 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the embodiment of the present application can be essentially or partly embodied in the form of a software product that contributes to the relevant technology. The software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0193] An embodiment of the present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, part or all of the steps in the information reminder method provided in the above embodiments are implemented.

[0194] An embodiment of the present application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute some or all of the steps in the information reminder method provided in the above embodiments.

[0195] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0196] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the embodiments of the present application may adopt the form of hardware embodiments, software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) that contain computer-usable program code.

[0197] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0198] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0199] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0200] The above description is merely an optional embodiment of the present application and is not intended to limit the protection scope of the present application.

Claims

1. An information reminder method, comprising: Determine at least one notification information obtained by the user terminal at a target time; Obtaining a user behavior spectrum within a specific time period before the target moment, wherein the user behavior spectrum is used to characterize the user's behavior activities with respect to at least one application in the user terminal within the specific time period; Based on the user behavior spectrum and the at least one notification information, target reminder information for reminding the user is determined.

2. The information reminder method according to claim 1, wherein the specific duration includes at least one time slice; The method further comprises: Determine user behavior information corresponding to each of the time slices, where the user behavior information is used to characterize the behavior activities of the user with respect to at least one application in the user terminal in each of the time slices; The user behavior spectrum is determined based on the user behavior information corresponding to each of the time slices.

3. The information reminder method according to claim 2, wherein determining the user behavior spectrum based on the user behavior information corresponding to each of the time slices comprises: Determine a first vector corresponding to the user behavior information corresponding to each of the time slices; The user behavior spectrogram is determined based on the first vectors respectively corresponding to the time slices.

4. The information reminder method according to claim 3, further comprising: Determining a target application triggered by the user in each of the time slices based on the user behavior information corresponding to each of the time slices; Determining text introduction information of the target application triggered by the user in each time slice; Determine a second vector corresponding to the text introduction information of the target application corresponding to each of the time slices; The determining the user behavior spectrum based on the first vectors respectively corresponding to the time slices includes: The user behavior spectrogram is determined based on the first vector and the second vector respectively corresponding to each of the time slices. 5 . The information reminder method according to claim 4 , wherein for different target applications having the same function, the second vectors corresponding to the text introduction information of the different target applications have similarity.

6. The information reminder method according to any one of claims 1 to 5, wherein the determining, based on the user behavior spectrum and the at least one notification information, target reminder information for reminding the user comprises: Determining a first target intention of the user based on the user behavior spectrum and the at least one notification information; Based on the first target intention, the target reminder information is determined.

7. The information reminder method according to claim 6, wherein determining the first target intention of the user based on the user behavior spectrum and the at least one notification information comprises: Based on the user behavior spectrum, determining at least one second target intention of the user; Based on the at least one second target intent and the at least one notification information, a first target intent of the user is determined.

8. The information reminder method according to claim 7, wherein determining the first target intention of the user based on the at least one second target intention and the at least one notification information comprises: Determining, based on the at least one second target intention, a confidence level corresponding to each of the at least one notification information; Based on the confidence levels respectively corresponding to the at least one notification information, the first target intention of the user is determined.

9. An information reminder device, comprising: A first determining module, used to determine at least one notification information obtained by the user terminal at a target time; An acquisition module, used to obtain a user behavior spectrum within a specific time period before the target moment, wherein the user behavior spectrum is used to characterize the user's behavior activities for at least one application in the user terminal within the specific time period; The second determination module is used to determine target reminder information for reminding the user based on the user behavior spectrum and the at least one notification information.

10. An electronic device comprising: A memory for storing executable data instructions; The processor is used to implement the information reminder method described in any one of claims 1 to 8 when executing the executable data instructions stored in the memory.

Images