Information recommendation method and device, equipment and medium

Abstract

The invention provides an information recommendation method and device, equipment and a medium, and the information recommendation method comprises the steps: obtaining first equipment information and equipment detection features of first equipment; according to the first equipment information, first habit information is obtained, and the first habit information comprises at least one first condition feature associated with the first equipment information and first operation information associated with the first condition feature; determining a first condition feature matched with the equipment detection feature from the at least one first condition feature, and taking the matched first condition feature as a target condition feature; and taking the first operation information associated with the target condition feature as recommended target operation information. The technical problem that in the prior art, recommendation of operation information is not accurate enough, and the intelligent effect of equipment is affected is solved.

Description

Technical Field

[0001] This disclosure relates to the field of Internet of Things (IoT) technology, and in particular to an information recommendation method, apparatus, device, and medium. Background Technology

[0002] In Internet of Things (IoT) applications, users can utilize IoT functionalities through a variety of devices. These devices include, for example, smart TVs, air conditioners, electronic devices, smart speakers, smart clothes racks, and so on. Different users may have different usage habits when using these devices. For instance, under certain conditions (such as someone moving or the temperature being too low), user A might perform operation A on device A, while user B might perform operation B on device B. Devices A and B can be the same device or different devices, and operations A and B may contain different operational information.

[0003] In related technologies, the recommended operational information is not accurate enough, which affects the intelligence effect of the equipment. Summary of the Invention

[0004] This disclosure aims to at least partially address one of the technical problems in the related art.

[0005] Therefore, this disclosure proposes an information recommendation method, apparatus, electronic device, computer-readable storage medium, and computer program product, which can improve the accuracy of operational information recommendation and enhance the intelligence of the equipment.

[0006] To achieve the above objectives, a first aspect of this disclosure provides an information recommendation method, comprising: acquiring first device information and device detection features of the first device; acquiring first habit information based on the first device information, wherein the first habit information includes: at least one first condition feature associated with the first device information and first operation information associated with the first condition feature; determining a first condition feature matching the device detection features from the at least one first condition feature, and using the matching first condition feature as a target condition feature; and using the first operation information associated with the target condition feature as the recommended target operation information.

[0007] To achieve the above objectives, a second aspect of this disclosure provides an information recommendation device, comprising: a first acquisition module for acquiring first device information and device detection features of the first device; a second acquisition module for acquiring first habit information based on the first device information, wherein the first habit information includes: at least one first condition feature associated with the first device information and first operation information associated with the first condition feature; a determination module for determining a first condition feature matching the device detection features from the at least one first condition feature, and using the matching first condition feature as a target condition feature; and a recommendation module for using the first operation information associated with the target condition feature as recommended target operation information.

[0008] To achieve the above objectives, a third aspect of this disclosure provides an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the information recommendation method proposed in the first aspect of this disclosure.

[0009] To achieve the above objectives, a fourth aspect of this disclosure provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the information recommendation method proposed in the first aspect of this disclosure.

[0010] To achieve the above objectives, a fifth aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the information recommendation method proposed in the first aspect of this disclosure.

[0011] To achieve the above objectives, a sixth aspect of this disclosure provides a processor for invoking computer execution instructions to perform the information recommendation method as described in the first aspect of this disclosure.

[0012] The information recommendation method, apparatus, electronic device, computer-readable storage medium, and computer program product disclosed herein obtain first device information and device detection features of the first device, and obtain first habit information based on the first device information. The first habit information includes at least one first conditional feature associated with the first device information and first operation information associated with the first conditional feature. A first conditional feature matching the device detection feature is determined from the at least one first conditional feature, and the matched first conditional feature is used as a target conditional feature. The first operation information associated with the target conditional feature is used as the recommended target operation information. Because determining the target operation information first involves obtaining suitable first habit information based on the device information (which may be pre-analyzed and learned) and containing one or more first conditional features and first operation information associated with each first conditional feature, and then matching a suitable target conditional feature from multiple first conditional features based on the device detection feature, and using the first operation information matching the target conditional feature as the recommended target operation information to the user, the accuracy of operation information recommendation and the intelligence effect of the device can be improved.

[0013] Additional aspects and advantages of this disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this disclosure. Attached Figure Description

[0014] The above and / or additional aspects and advantages of this disclosure will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, in which:

[0015] Figure 1 This is a flowchart illustrating an information recommendation method provided in an embodiment of the present disclosure;

[0016] Figure 2 A flowchart illustrating another information recommendation method provided in this embodiment of the disclosure;

[0017] Figure 3 This is a schematic diagram of the candidate behavior template in an embodiment of this disclosure;

[0018] Figure 4 A flowchart illustrating another information recommendation method provided in this embodiment of the disclosure;

[0019] Figure 5 This is a schematic diagram illustrating the input of candidate habit information into the database in an embodiment of this disclosure;

[0020] Figure 6 This is a schematic diagram of the customary reach process in the embodiments of this disclosure;

[0021] Figure 7This is a schematic diagram of the recommended system architecture in an embodiment of this disclosure;

[0022] Figure 8 This is a schematic diagram of the structure of an information recommendation device provided in an embodiment of the present disclosure;

[0023] Figure 9 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. Detailed Implementation

[0024] Some embodiments of this disclosure will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. Various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent upon understanding this disclosure. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but can be changed as will become apparent upon understanding this disclosure, except for operations that must be performed in a particular order. Furthermore, for clarity and brevity, descriptions of features known in the art may be omitted.

[0025] The embodiments described in the following examples of this disclosure are not representative of all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0026] It should be noted that all actions involving the acquisition of signals, information, or data in this disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with authorization from the owner of the relevant device. Figure 1 This is a flowchart illustrating an information recommendation method provided in an embodiment of this disclosure.

[0027] The information recommendation method and apparatus of this disclosure are described below with reference to the accompanying drawings.

[0028] Figure 1 This is a flowchart illustrating an information recommendation method provided in an embodiment of this disclosure.

[0029] This embodiment illustrates the example of an information recommendation method configured in an information recommendation device. In this embodiment, the information recommendation method can be configured in an information recommendation device, which can be located in a server or in an electronic device, without limitation.

[0030] This embodiment uses an information recommendation method configured in an electronic device as an example. The electronic device includes hardware devices with various operating systems, such as smartphones, tablets, personal digital assistants, and e-readers.

[0031] It should be noted that the execution subject of the embodiments of this disclosure may be, in hardware, an image sensor and / or an image signal processor (ISP) in a server or electronic device, or an image processor, and in software, a related background service in a server or electronic device, and there is no limitation thereto.

[0032] like Figure 1 As shown, this information recommendation method includes the following steps:

[0033] Step S101: Obtain the first device information and the device detection features of the first device.

[0034] The device currently being used by the user (such as a smart TV, air conditioner remote control, mobile phone, smart washing machine, smart door lock, etc.), or a device that automatically detects scene features (such as temperature detection device, light detection device, etc.), can be referred to as the first device. First device information describes the first device. This information may include, for example, the device's identifier, serial number, and device type.

[0035] The device detection features can refer to the features detected by the first device. These features can be used to describe object behavior (e.g., user behavior); and / or, they can be used to describe a scenario.

[0036] For example, if the first device is the device currently being used by the user, then the first device can detect user behavior characteristics. For example, user behavior characteristics detected by a smart washing machine (which control button the user triggered, the washing mode set by the user, etc.), or user behavior characteristics detected by a smart door lock (such as the smart door lock detecting that the user opened the door and detected that the user entered the room from the outside, etc.), etc., without any restrictions.

[0037] For example, if the first device is a device that automatically detects scene features, then the first device can detect scene features. For example, scene features (ambient temperature) detected by a temperature detection device, or scene features (such as light intensity) detected by a light detection device, etc., without limitation.

[0038] Step S102: Obtain first habit information based on the first device information, wherein the first habit information includes: at least one first condition feature associated with the first device information, and first operation information associated with the first condition feature.

[0039] Among them, habit information can be used to describe a user's usage habits when using at least one device simultaneously. First habit information refers to candidate habit information that matches the first device information.

[0040] Optionally, in some embodiments, the first habit information can be obtained based on the first device information using artificial intelligence; or the habit information pre-configured for the first device can be obtained as the first habit information; or any other possible method can be used to obtain the first habit information based on the first device information, without limitation.

[0041] The first conditional feature is, for example, a feature in the first habitual information used to trigger the operation information. The operation information is, for example, the various operations specifically described in the first habitual information.

[0042] For example, the first device information is the information of a temperature detection device. The first habitual information matching the first device information is "turn on the air conditioner to cool when the temperature is above 30 degrees". Therefore, the first conditional feature is "the temperature is above 30 degrees", and the first operation information associated with the first conditional feature is "turn on the air conditioner to cool".

[0043] For example, the first device information is information about a smart washing machine. The first habitual information that matches the first device information is "lower the clothes rack when washing is finished", then the first conditional feature is "washing is finished", and the first operation information associated with the first conditional feature is "lower the clothes rack".

[0044] For example, the first device information is information from a mobile phone. The first habitual information that matches the first device information is "turn on the corridor light when someone moves". Then the first conditional feature is "someone moves", and the first operation information associated with the first conditional feature is "turn on the corridor light".

[0045] The examples above are merely illustrations. In practice, the first conditional features in the first habitual information and the first operational information associated with the first conditional features can be personalized for learning modeling or configuration according to actual usage needs. There are no restrictions on this.

[0046] Optionally, in some embodiments, candidate habit information that matches the first device information can be selected from multiple candidate habit information, and the candidate habit information that matches the first device information can be used as the first habit information.

[0047] Optionally, in some embodiments, in the process of obtaining first habit information based on first device information, multiple candidate habit information may be obtained. These candidate habit information include: candidate device information, candidate condition features associated with the candidate device information, and candidate operation information associated with the candidate condition features. Candidate device information identical to the first device information is selected from the multiple candidate device information, and the candidate habit information to which the candidate device information identical to the first device information belongs is taken as the first habit information. The candidate condition features in the belonging candidate habit information are taken as the first condition features, and the candidate operation information in the belonging candidate habit information is taken as the first operation information. This improves the efficiency of obtaining first habit information and supports the acquisition of accurate first habit information. When recommending target operation information to the user based on the first habit information, the recommendation accuracy can be greatly improved, enhancing the user experience.

[0048] The candidate habit information can be habit information learned in advance based on user behavior data analysis. For example, candidate habit information could be such as "turn on the air conditioner when the temperature is above 30 degrees Celsius", "lower the clothes hanger when the laundry is finished", "turn on the hallway light when someone moves", and so on.

[0049] In other words, in this embodiment of the present disclosure, candidate habit information matching the first device information can be detected from multiple candidate habit information and used as the first habit information. Since the different candidate habit information may be associated with different candidate devices, each candidate device has corresponding candidate device information. For example, for the candidate habit information "turn on the air conditioner when the temperature is above 30 degrees Celsius," the associated candidate device may be a temperature detection device; for the candidate habit information "lower the clothes hanger when washing is finished," the associated candidate device may be a smart washing machine; for the candidate habit information "turn on the corridor light when someone moves," the associated candidate device may be a mobile phone. Therefore, it is possible to first select candidate device information that is identical to the first device information, determine the candidate habit information to which this "identical candidate device information" belongs, and then use the candidate habit information matching the first device information as the first habit information.

[0050] Step S103: Determine a first condition feature that matches the device detection feature from at least one first condition feature, and use the matching first condition feature as the target condition feature.

[0051] The aforementioned first habitual information may include at least one first conditional feature associated with the first device information, and first operation information associated with the first conditional feature. Therefore, in this embodiment of the disclosure, a first conditional feature matching the device detection feature can be determined from at least one first conditional feature. Here, "device detection feature matches first conditional feature" can be understood as the device detection feature conforming to the first conditional feature. If the device detection feature is used to describe object behavior, then "device detection feature matches first conditional feature" can indicate that the object behavior conforms to the behavioral characteristics described by the first conditional feature. If the device detection feature is used to describe scene characteristics, then "device detection feature matches first conditional feature" can indicate that the scene characteristics conform to the scene characteristics described by the first conditional feature.

[0052] After determining the first condition feature that matches the device detection feature from at least one first condition feature, the matching first condition feature can be used as the target condition feature.

[0053] For example, taking the information of a temperature detection device as the first device information, the first habitual information could be, for instance, "If the temperature is above 30 degrees, turn on the air conditioner to cool and humidify," or "If the temperature is above 25 degrees, turn on the air conditioner to cool," or any other possible habitual information. Therefore, the first conditional feature could be either "temperature above 30 degrees" or "temperature above 25 degrees," meaning the first device information is associated with two first conditional features. The first operational information associated with the first device information and each first conditional feature could be two different first operational pieces of information: "turn on the air conditioner to cool and humidify" and "turn on the air conditioner to cool."

[0054] Therefore, in this embodiment of the disclosure, a suitable first condition feature can also be matched based on the device detection features. For example, the current scene feature detected by the temperature detection device can be used to determine whether it is "temperature higher than 30 degrees" or "temperature higher than 25 degrees", thereby further matching an accurate first condition feature (the matched first condition feature can be called the target condition feature). If the scene feature detected by the temperature detection device is a temperature of 26 degrees, then the matched first condition feature (target condition feature) is determined to be "temperature higher than 25 degrees". If the scene feature detected by the temperature detection device is a temperature of 35 degrees, then the matched first condition feature (target condition feature) is determined to be "temperature higher than 30 degrees". There is no limitation on this.

[0055] Step S104: The first operation information associated with the target condition feature is used as the recommended target operation information.

[0056] The above-mentioned method involves determining a first condition feature that matches the device detection feature from at least one first condition feature in the first habit information, and using the matched first condition feature as the target condition feature. It may also use the first operation information associated with the target condition feature in the first habit information as the recommended target operation information.

[0057] The first operation information that matches the first device information and the device detection features of the first device can be referred to as the target operation information. Operation information can be used to describe suggested device operations. Examples of operation information include suggestions such as "turn on the air conditioner to cool" or "lower the clothes hanger." The first operation information refers to the operation information associated with the first device information and the first conditional features, obtained through pre-learning modeling in the first habit information. The target operation information can be the first operation information matched with the first device information and the target conditional features.

[0058] For example, taking the information of a temperature detection device as the first device information, the first habitual information could be, for example, "If the temperature is above 30 degrees, turn on the air conditioner to cool and humidify." Based on the device detection characteristics, the target condition characteristic "temperature above 30 degrees" is determined. The first operation information corresponding to the information of the temperature detection device and the target condition characteristic "temperature above 30 degrees" is "turn on the air conditioner to cool and humidify." Therefore, "turn on the air conditioner to cool and humidify" can be recommended as the target operation information to the user.

[0059] In this embodiment, by acquiring first device information and device detection features of the first device, and based on the first device information, first habit information is acquired. The first habit information includes at least one first conditional feature associated with the first device information and first operation information associated with the first conditional feature. A first conditional feature matching the device detection feature is determined from the at least one first conditional feature, and the matched first conditional feature is used as the target conditional feature. The first operation information associated with the target conditional feature is used as the recommended target operation information. Since, when determining the target operation information, suitable first habit information is first acquired based on the device information (which may be pre-analyzed and learned), and the first habit information includes one or more first conditional features and first operation information associated with each first conditional feature, and then a suitable target conditional feature is matched from multiple first conditional features based on the device detection feature, and the first operation information matching the target conditional feature is used as the target operation information recommended to the user, the accuracy of operation information recommendation can be improved, and the intelligence effect of the device can be enhanced.

[0060] Optionally, in some embodiments of this disclosure, the first conditional feature includes: behavioral features and / or scene features, wherein the behavioral features include features of the object's behavioral dimension, and the scene features include features of the scene dimension.

[0061] For example, the first conditional feature can include behavioral features. Behavioral features include, for example, user behavior-related features such as "going home," "sleeping," "taking a shower," and "someone moving." The first conditional feature can also include scene features. Scene features include, for example, scene-related features such as "temperature above 30 degrees," "temperature above 25 degrees," and "laundry finished." The first conditional feature can also include both behavioral and scene features.

[0062] Optionally, in some embodiments of this disclosure, in the process of determining a first condition feature that matches a device detection feature from at least one first condition feature and using the matching first condition feature as a target condition feature, a behavioral feature that matches a device detection feature may be selected from at least one behavioral feature, wherein the device detection feature is used to describe object behavior; wherein the matching behavioral feature may be used as the target condition feature.

[0063] For example, “behavioral features that match device detection features”, such as: if a smart door lock detects that the door lock is opened from the outside, then the behavioral feature that matches the device detection feature is “coming home”; or if a smart shower detects that a continuous spraying operation, then the behavioral feature that matches the device detection feature is “taking a shower”.

[0064] Optionally, in some embodiments of this disclosure, in the process of determining a first condition feature that matches the device detection feature from at least one first condition feature and using the matching first condition feature as the target condition feature, the process may involve selecting a scene feature that matches the device detection feature from at least one scene feature, wherein the device detection feature is used to describe the scene; wherein the matching scene feature may be used as the target condition feature.

[0065] For example, "scene features that match the device detection features" means that if the device detection feature is: the temperature detection device detects a temperature of 26 degrees, then the scene feature that matches the device detection feature is "temperature higher than 25 degrees". Or, if the device detection feature is: the temperature detection device detects a temperature of 31 degrees, then the scene feature that matches the device detection feature is "temperature higher than 30 degrees".

[0066] Optionally, in some embodiments of this disclosure, in the process of determining a first condition feature that matches the device detection feature from at least one first condition feature and using the matching first condition feature as the target condition feature, it is also possible to select a behavioral feature that matches the device detection feature from at least one behavioral feature and to select a scene feature that matches the device detection feature from at least one scene feature, and use the matching behavioral feature and the matching scene feature together as the target condition feature.

[0067] This can effectively improve the accuracy of target condition feature detection, significantly improve the accuracy of the first operation information recommendation, and enhance the intelligence of the equipment.

[0068] Figure 2 This is a schematic flowchart of another information recommendation method provided in an embodiment of this disclosure.

[0069] like Figure 2 As shown, this information recommendation method includes the following steps:

[0070] Step S201: Obtain multiple candidate behavior templates and multiple object behavior data. The candidate behavior templates include: trigger features, template condition features corresponding to the trigger features, and execution information corresponding to the template condition features.

[0071] In this embodiment of the disclosure, candidate habit information can be generated by referring to multiple candidate behavior templates and multiple object behavior data.

[0072] The candidate behavior template includes: a trigger feature, a corresponding template condition feature, and execution information corresponding to the template condition feature. The trigger feature can refer to the detection features related to the device that triggers template condition feature matching. For example, the trigger feature could be a feature detected by the device that triggers template condition feature matching, along with device information. The corresponding template condition feature could be a feature configured for each template condition. The execution information can be the corresponding operation information pre-configured for the trigger feature and the template condition feature.

[0073] like Figure 3 , Figure 3 This is a schematic diagram of candidate behavior templates in an embodiment of this disclosure. Multiple candidate behavior templates may include "turn on the air conditioner when the temperature is high," "turn on the lights when someone moves and the ambient light is dim," and "automatically do some things for you when you get home." Triggering features include, for example, detection features needed to determine "high temperature," detection features needed to determine "whether someone is present" and "ambient light," and detection features needed to determine whether you have "returned home." Template condition features include, for example, "high temperature," "someone moves," "dim ambient light," and "returned home." Execution information includes, for example, "turn on the air conditioner," "turn on the lights," and "automatically do some things for you," without limitation. After collecting multiple candidate behavior templates, multiple object behavior data (e.g., user behavior data) can be collected. Then, user habits (an optional example of candidate habit information) are generated based on the multiple candidate behavior templates and multiple user behavior data.

[0074] Step S202: Based on multiple candidate behavior templates and multiple object behavior data, candidate habit information is generated, wherein the candidate habit information includes: candidate device information, candidate condition features associated with the candidate device information, and candidate operation information associated with the candidate condition features.

[0075] After obtaining multiple candidate behavior templates and multiple object behavior data, artificial intelligence can be used to model and analyze the multiple candidate behavior templates and multiple object behavior data to obtain one or more candidate habit information; or statistical analysis can be used to statistically analyze the multiple candidate behavior templates and multiple object behavior data to generate candidate habit information; or any other possible method can be used to process the multiple candidate behavior templates and multiple object behavior data to generate candidate habit information.

[0076] Optionally, when generating candidate habit information, a target behavior template can be selected from multiple candidate behavior templates based on object behavior data, and frequently occurring object behavior data can be integrated into the target behavior template to generate candidate habit information. There are no restrictions on this.

[0077] For example, if the target behavior template is "turn on the air conditioner when the temperature is high", and based on multiple object behavior data, it is determined that the most frequent occurrence is that users will turn on the air conditioner when the temperature is above 30 degrees Celsius, then "turn on the air conditioner when the temperature is above 30 degrees Celsius" can be used as candidate habit information.

[0078] Step S203: Obtain the first device information and the device detection features of the first device.

[0079] Step S204: Select candidate device information that is the same as the first device information from multiple candidate device information, and take the candidate habit information to which the candidate device information that is the same as the first device information belongs as the first habit information.

[0080] Step S205: Take the candidate condition features in the candidate habit information as the first condition features, and take the candidate operation information in the candidate habit information as the first operation information.

[0081] Step S206: Determine a first condition feature that matches the device detection feature from at least one first condition feature, and use the matching first condition feature as the target condition feature.

[0082] Step S207: The first operation information associated with the target condition feature is used as the recommended target operation information.

[0083] For a detailed description of S203-S207, please refer to the above embodiments, which will not be repeated here.

[0084] In this embodiment, by acquiring first device information and device detection features of the first device, and based on the first device information, first habit information is acquired. The first habit information includes at least one first conditional feature associated with the first device information and first operation information associated with the first conditional feature. A first conditional feature matching the device detection feature is determined from the at least one first conditional feature, and the matched first conditional feature is used as the target conditional feature. The first operation information associated with the target conditional feature is used as the recommended target operation information. Since, when determining the target operation information, suitable first habit information is first acquired based on the device information (which may be pre-analyzed and learned), and the first habit information includes one or more first conditional features and first operation information associated with each first conditional feature, and then a suitable target conditional feature is matched from multiple first conditional features based on the device detection feature, and the first operation information matching the target conditional feature is used as the target operation information recommended to the user, the accuracy of operation information recommendation can be improved, and the intelligence effect of the device can be enhanced. By acquiring multiple candidate behavior templates and multiple object behavior data, where the candidate behavior templates include: trigger features, template condition features corresponding to the trigger features, and execution information corresponding to the template condition features, candidate habit information is generated based on multiple candidate behavior templates and multiple object behavior data. Since the candidate habit information is based on the analysis and learning of multiple object behavior data, it supports the improvement of recommendation accuracy. Furthermore, the analysis and learning process also incorporates multiple candidate behavior templates, which can further improve the generation efficiency and accuracy of candidate habit information.

[0085] Figure 4 This is a schematic flowchart of another information recommendation method provided in an embodiment of this disclosure.

[0086] like Figure 4 As shown, this information recommendation method includes the following steps:

[0087] Step S401: Obtain multiple candidate behavior templates and multiple object behavior data. The candidate behavior templates include: trigger features, template condition features corresponding to the trigger features, and execution information corresponding to the template condition features.

[0088] For a detailed description of S401, please refer to the above embodiments, which will not be repeated here.

[0089] The definition of a candidate behavior template can be described as follows:

[0090] Triggering features can be represented as "trigger"; template condition features can be represented as "condition"; and execution information can be represented as "action". A set of candidate behavior templates can be represented as TCA. There can be one or more triggering features, and multiple triggers can have the same meaning. There can be one or more template condition features, or zero; multiple conditions can have the same meaning. There can be one or more execution information actions.

[0091] The candidate behavior templates can be sourced from various sources, including: manual definition, automated data mining, and combinations obtained from raw user device logs using time-series clustering algorithms. The reasonableness of the relationships between the various elements in the candidate behavior template can be determined through pre-screening for reasonableness. All elements in the candidate behavior template must have reasonable relationships with each other.

[0092] Step S402: Based on multiple object behavior data and candidate behavior templates, determine the statistical reference value corresponding to each statistical dimension. The statistical reference value is used to quantitatively describe the degree of matching between multiple object behavior data and candidate behavior templates under the statistical dimension.

[0093] The statistical dimensions can be, for example, time-based (e.g., time slices or days) or frequency-based. Multiple object behavior data can be object behavior data for 20 consecutive working days, 8 consecutive weekend days, or 28 consecutive weekend days.

[0094] After obtaining multiple object behavior data, the above can be combined with candidate behavior templates to determine the statistical reference value corresponding to each statistical dimension. The statistical reference value is used to quantitatively describe the degree of matching between multiple object behavior data and candidate behavior templates under the statistical dimension.

[0095] In this embodiment of the disclosure, different priorities can also be configured for different statistical dimensions. The priority of the time slice dimension can be higher than the priority of the day dimension, and the priority of the day dimension can be higher than the priority of the count ratio dimension. There are no restrictions on this.

[0096] Optionally, in some embodiments of this disclosure, based on each statistical dimension, if the current object behavior data matches the first part of the candidate behavior template, a first target value is obtained by accumulating a preset value on the first value, wherein the first value is determined based on the previous object behavior data; based on each statistical dimension, if the current object behavior data matches the second part of the candidate behavior template, a second target value is obtained by accumulating a preset value on the second value, wherein the second value is determined based on the previous object behavior data; based on the first target value, the second target value, and the statistical information corresponding to each statistical dimension, a statistical reference value corresponding to each statistical dimension is determined. This effectively improves the reference value of the statistical reference value, and when filtering target behavior templates based on the statistical reference value, it can significantly improve the accuracy of target behavior template selection.

[0097] The default value is, for example, 1.

[0098] In other words, multiple object behavior data can be traversed. Each time object behavior data is traversed, it's determined whether the object behavior data matches a portion of the candidate behavior template. "Matching" means determining that a portion of the candidate behavior template has occurred based on the object behavior data. For example, if the candidate behavior template is "turn on the air conditioner when the temperature is high," and based on a certain object behavior data, it's determined that the object turned on the air conditioner when the temperature was above 30 degrees Celsius, then the object behavior data matches the candidate behavior template. The matched portion includes: temperature detection feature, high temperature, and turning on the air conditioner. This matched portion can be called the first part of the content. On the other hand, if the matched portion only includes: temperature detection feature and high temperature (indicating that the user took some action when the temperature was high, but not necessarily turning on the air conditioner; it could also be opening a window), then only the second part of the content has been matched.

[0099] In other words, the first part includes: trigger features, template condition features, and execution information; the second part includes: trigger features and template condition features. In the embodiments of this disclosure, "the object behavior data and the first part of the candidate behavior template match" can also be referred to as "TCA occurs", and "the object behavior data and the second part of the candidate behavior template match" can also be referred to as "TC occurs", without limitation.

[0100] When iterating through object behavior data, if the current object behavior data matches the first part of the candidate behavior template, the first value is incremented by 1 to obtain the first target value. The first value is determined based on the previous object behavior data. The method of determining the first value based on the previous object behavior data can be based on the processing method of the current object behavior data, and there are no restrictions on this.

[0101] When iterating through object behavior data, if the current object behavior data matches the second part of the candidate behavior template, the second value is incremented by 1 to obtain the second target value. The second value is determined based on the previous object behavior data. The method for determining the second value based on the previous object behavior data can be based on the processing method of the current object behavior data, and there are no restrictions on this.

[0102] After obtaining the first and second target values ​​for each statistical dimension from the above statistics, statistical reference values ​​for each statistical dimension can be determined based on the first and second target values ​​and the statistical information corresponding to each statistical dimension, without any restrictions.

[0103] The statistical information can include, for example, the statistical period, the number of days in the statistical period, etc., and there are no restrictions on this.

[0104] Optionally, in some embodiments, the process of determining the statistical reference value corresponding to each statistical dimension based on the first target value, the second target value, and the statistical information corresponding to each statistical dimension may involve determining the first descriptive value corresponding to each statistical dimension based on the first and second target values, determining the second descriptive value corresponding to each statistical dimension based on the second target value and the statistical information corresponding to each statistical dimension, and determining the statistical reference value corresponding to each statistical dimension based on the first and second descriptive values. This effectively improves the reference value of the statistical reference value, and significantly enhances the accuracy of target behavior template selection when filtering target behavior templates based on the statistical reference value.

[0105] Here, the first target value can also be referred to as the numerator count, and the second target value can also be referred to as the denominator count. The first descriptive value can be the ratio of the first target value to the second target value. The second descriptive value can be the ratio of the second target value to statistical information. The first descriptive value can also be referred to as the occurrence ratio, and the second descriptive value can also be referred to as the count ratio or the number of days it occurs.

[0106] Optionally, in some embodiments, in the process of determining the statistical reference value corresponding to each statistical dimension based on the first description value and the second description value, the first description value and the second description value can be used together as the statistical reference value, or a confidence level can be calculated based on the first description value and the second description value, and the calculated confidence level can be used as the statistical reference value under the corresponding statistical dimension. There is no limitation on this.

[0107] Examples illustrating the above description can be as follows:

[0108] Mining strategies can include mining by time slice, mining by day, and mining by frequency ratio. Mining by time slice is an optional example of statistical analysis by time slice; mining by day is an optional example of statistical analysis by day; and mining by frequency ratio is an optional example of statistical analysis by frequency ratio. There are no restrictions on these strategies.

[0109] Mining based on time slices (a day of 24 hours can be divided into 12 time slices) includes:

[0110] The number of TCA occurrences exceeds the preset number and "days + time slices" appear consecutively.

[0111] Molecular count: If a TCA that satisfies the template condition features appears within a time slice (i.e., an optional example where the object behavior data and the first part of the candidate behavior template match in terms of time slice dimension), then the molecular count for that day + time slice is incremented by 1.

[0112] Denominator count: If a TC that meets the template condition features appears within a time slice (i.e., an optional example where the object behavior data and the second part of the candidate behavior template match in terms of time slice dimension), then the denominator count for that day + time slice is increased by 1.

[0113] The confidence score is calculated within the same time slice dimension within the calculation period (which can also be called the statistical period, an optional example of the above statistical information) to obtain the confidence score (an optional example of the statistical reference value).

[0114] The extraction logic may include:

[0115] Occurrence rate: Number of numerator counts / Number of denominator counts ≥ 50%;

[0116] Count ratio: Number of counts in the denominator / statistical period ≥ 50%;

[0117] Confidence level: Number of TCA occurrences / Number of TC occurrences.

[0118] In other words, candidate behavior templates are extracted that correspond to a ratio of numerator counts / denominator counts ≥ 50%, and candidate behavior templates that correspond to a ratio of denominator counts / statistical period ≥ 50%. The number of TCA occurrences and TC occurrences corresponding to the candidate behavior templates are counted. Then, the ratio of the number of TCA occurrences and the number of TC occurrences is used as the confidence level of "mining by daily dimension" (an optional example of statistical reference value).

[0119] Mining based on daily dimensions includes:

[0120] The number of TCA occurrences exceeds the preset number for consecutive days.

[0121] Number of occurrences: If a TCA that meets the template condition characteristics appears on a given day (i.e., an optional example where the object behavior data and the first part of the candidate behavior template match on a daily basis), then increment the count for that day by 1.

[0122] The number of occurrences in the denominator is incremented by 1 if a TC (i.e., an optional example where the object behavior data and the second part of the candidate behavior template match on a daily basis) appears on the same day.

[0123] The confidence level is calculated for the same day dimension within the calculation period (which can also be called the statistical period, an optional example of the above statistical information) to obtain the confidence level (an optional example of the statistical reference value).

[0124] The extraction logic may include:

[0125] Occurrence rate: Number of numerator counts / Number of denominator counts ≥ 50%;

[0126] Count ratio: Number of counts in the denominator / statistical period ≥ 50%;

[0127] Confidence level = Number of TCA occurrences / Number of TC occurrences.

[0128] In other words, candidate behavior templates are extracted that correspond to a ratio of numerator counts / denominator counts ≥ 50%, and candidate behavior templates that correspond to a ratio of denominator counts / statistical period ≥ 50%. The number of TCA occurrences and TC occurrences corresponding to the candidate behavior templates are counted. Then, the ratio of the number of TCA occurrences and the number of TC occurrences is used as the confidence level of "mining by daily dimension" (an optional example of statistical reference value).

[0129] Mining based on the proportion of times includes:

[0130] Number of occurrences of TCA (i.e., an optional example where the object behavior data and the first part of the candidate behavior template match) within the calculation period;

[0131] Denominator count: The number of times TC appears within the calculation period (i.e., an optional example where the object behavior data and the second part of the candidate behavior template match).

[0132] The extraction logic may include:

[0133] Occurrence rate: Number of numerator counts / Number of denominator counts ≥ 50%;

[0134] Days of occurrence: The ratio of days TCA occurred to the total number of days counted is ≥20%;

[0135] Confidence level = Number of TCA occurrences / Number of TC occurrences.

[0136] In other words, candidate behavior templates are extracted that correspond to a ratio of numerator counts to denominator counts of ≥50%, and candidate behavior templates that correspond to a ratio of TCA occurrences to statistical days of ≥20%. The number of TCA occurrences and TC occurrences corresponding to the candidate behavior templates are counted. Then, the ratio of the number of TCA occurrences to the number of TC occurrences is used as the confidence level of "mining by daily dimension" (an optional example of statistical reference value).

[0137] Step S403: Select the target behavior template corresponding to the statistical dimension from multiple candidate behavior templates based on the statistical reference value.

[0138] After obtaining the statistical reference value corresponding to each statistical dimension, the target behavior template corresponding to the statistical dimension can be selected from multiple candidate behavior templates based on the statistical reference value.

[0139] Optionally, in some embodiments, in the process of selecting a target behavior template corresponding to a statistical dimension from multiple candidate behavior templates based on statistical reference values, a statistical reference value greater than a threshold may be selected from multiple statistical reference values, and the candidate behavior template corresponding to the selected statistical reference value may be used as the target behavior template. The number of candidate behavior templates corresponding to the selected statistical reference value may be one or more. The number of selected statistical reference values ​​may also be one or more.

[0140] Step S404: Generate candidate habit information based on the target behavior template.

[0141] When generating candidate habit information based on target behavior templates, the above process can involve aggregating at least one target behavior template to obtain candidate habit information. The aggregation strategy can be as follows:

[0142] TCAs with the same T, C, time_window (e.g., statistical period, statistical days), template identifier (used to identify the target behavior template), and confidence interval (confidence interval: 0.3~0.5, 0.5~0.8, 0.8~1.0) can be aggregated into a unique candidate habit information, without any restrictions.

[0143] In addition, confidence levels can be assigned to candidate habit information. The confidence level of the aggregated candidate habit information is the lowest confidence level among the confidence levels corresponding to multiple target behavior templates.

[0144] Optionally, in some embodiments, the process of generating candidate habit information based on a target behavior template may involve generating candidate condition features based on template condition features and at least some object behavior data in the target behavior template, determining candidate device information based on trigger features in the target behavior template, generating candidate operation information based on execution information and at least some object behavior data in the target behavior template, and generating candidate habit information based on candidate condition features, candidate device information, and candidate operation information. This further improves the accuracy of candidate habit information generation and can significantly support improving the accuracy of operation information recommendations.

[0145] Step S405: Obtain the first device information and the device detection features of the first device.

[0146] Step S406: Select candidate device information that is the same as the first device information from multiple candidate device information, and take the candidate habit information to which the candidate device information that is the same as the first device information belongs as the first habit information; wherein, the candidate condition features in the candidate habit information to which it belongs are taken as the first condition features, and the candidate operation information in the candidate habit information to which it belongs are taken as the first operation information.

[0147] Step S407: Determine a first condition feature that matches the device detection feature from at least one first condition feature, and use the matching first condition feature as the target condition feature.

[0148] Step S408: The first operation information associated with the target condition feature is used as the recommended target operation information.

[0149] For a detailed description of S405-S408, please refer to the above embodiments, which will not be repeated here.

[0150] In this embodiment, by acquiring first device information and device detection features of the first device, and based on the first device information, first habit information is acquired. The first habit information includes at least one first conditional feature associated with the first device information and first operation information associated with the first conditional feature. A first conditional feature matching the device detection feature is determined from the at least one first conditional feature, and the matched first conditional feature is used as the target conditional feature. The first operation information associated with the target conditional feature is used as the recommended target operation information. Since, when determining the target operation information, suitable first habit information is first acquired based on the device information (which may be pre-analyzed and learned), and the first habit information includes one or more first conditional features and first operation information associated with each first conditional feature, and then a suitable target conditional feature is matched from multiple first conditional features based on the device detection feature, and the first operation information matching the target conditional feature is used as the target operation information recommended to the user, the accuracy of operation information recommendation can be improved, and the intelligence effect of the device can be enhanced. By analyzing multiple object behavior data and candidate behavior templates, a statistical reference value is determined for each statistical dimension. This reference value quantifies the degree of match between the object behavior data and candidate behavior templates within that statistical dimension. Based on the reference value, a target behavior template corresponding to the statistical dimension is selected from the candidate behavior templates, and candidate habit information is generated from the target behavior template. This effectively enhances the reference value of the statistical reference value, significantly improving the accuracy of target behavior template selection when filtering based on it. Furthermore, generating candidate habit information based on the target behavior template further improves the accuracy of the generated candidate habit information.

[0151] like Figure 5 As shown, Figure 5 This is a schematic diagram illustrating the input of candidate habit information into the database in an embodiment of this disclosure. Figure 5 In this system, multiple candidate habit information can be obtained based on habit algorithm analysis. These candidate habits can then be processed for habit storage. Furthermore, optimization processing of these candidate habits can be performed, including: confidence level control, habit deduplication, and similarity handling. Specifically, confidence level control allows only high-confidence candidate habits to be retained. Habit deduplication separates candidate habits to be stored from those already stored. Similarity handling performs conflict detection or aggregation between candidate habits to be stored and those already stored. Finally, in terms of user-side perception, recommendations can be automatically made to users when habit recommendation is enabled.

[0152] like Figure 6 As shown, Figure 6This is a schematic diagram of the habit outreach process in this embodiment. Contextual features can be extracted, and data can be persisted to the server disk by combining domain services and user profile services (i.e., forming candidate habit information and storing it on the server). During this process, object behavior data can be extracted, using the object as an example of a user. Object behavior data can include: a basic user profile, a user device / attribute operation sequence (T-1), where T-1 represents several historical time points, T is a positive integer greater than 1, and the user device / attribute operation sequence is real-time. After forming the candidate habit information, a recommendation model can be trained. For example, the candidate habit information and object behavior data can be used as raw samples, and then training samples can be extracted and used to train the recommendation model to provide online recommendation services.

[0153] like Figure 7 As shown, Figure 7 This is a schematic diagram of the recommended system architecture in an embodiment of this disclosure. Figure 7 In the online phase, habits can be recommended to users (the recommended habits can include the target operation information mentioned above, as well as the first habit information to which the target operation information belongs). In the offline phase, candidate habit information can be formed based on multiple candidate behavior templates (TCAs), and the candidate habit information can be stored in the database.

[0154] This embodiment supports the definition of behavior templates. Based on automated settings and object behavior data, and through a large model's rationality judgment and manual supplementation, behavior templates can be defined. Combinations of behavior templates are used to mine object habit information. The probability of an object periodically executing a certain behavior template is calculated based on object behavior data and behavior templates across different statistical dimensions. Behavior templates with high probability are defined as habit information after rationality processing. The mined habit information can also be merged, conflict detected, updated, and expired to ensure the real-time performance and effectiveness of object-oriented habit recommendations. Furthermore, it supports recommending habits to objects at multiple times, merging and sorting various habits before displaying them to the object to capture the object's most authentic usage needs.

[0155] Figure 8 This is a schematic diagram of the structure of an information recommendation device provided in an embodiment of this disclosure.

[0156] like Figure 8 As shown, the information recommendation device 80 includes:

[0157] The first acquisition module 801 is used to acquire the first device information and the device detection features of the first device.

[0158] The second acquisition module 802 is used to acquire first habit information based on the first device information, wherein the first habit information includes: at least one first condition feature associated with the first device information, and first operation information associated with the first condition feature.

[0159] The determination module 803 is used to determine a first condition feature that matches the device detection feature from at least one first condition feature, and to use the matching first condition feature as the target condition feature.

[0160] The recommendation module 804 is used to use the first operation information associated with the target condition feature as the recommended target operation information.

[0161] It should be noted that the foregoing explanation of the information recommendation method also applies to the information recommendation device of this embodiment, and will not be repeated here.

[0162] In this embodiment, by acquiring first device information and device detection features of the first device, and based on the first device information, first habit information is acquired. The first habit information includes at least one first conditional feature associated with the first device information and first operation information associated with the first conditional feature. A first conditional feature matching the device detection feature is determined from the at least one first conditional feature, and the matched first conditional feature is used as the target conditional feature. The first operation information associated with the target conditional feature is used as the recommended target operation information. Since, when determining the target operation information, suitable first habit information is first acquired based on the device information (which may be pre-analyzed and learned), and the first habit information includes one or more first conditional features and first operation information associated with each first conditional feature, and then a suitable target conditional feature is matched from multiple first conditional features based on the device detection feature, and the first operation information matching the target conditional feature is used as the target operation information recommended to the user, the accuracy of operation information recommendation can be improved, and the intelligence effect of the device can be enhanced.

[0163] Figure 9 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. Figure 9 The electronic device 12 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0164] like Figure 9 As shown, the electronic device 12 is represented in the form of a general-purpose computing device. The components of the electronic device 12 may include, but are not limited to: one or more processors or processing units 16, memory 28, and bus 18 connecting different system components (including memory 28 and processing unit 16).

[0165] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0166] Electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 12, including volatile and non-volatile media, removable and non-removable media.

[0167] Memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 30 and / or cache 32. Electronic device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (… Figure 9 Not shown; usually referred to as a "hard drive".

[0168] although Figure 9 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disc drive for reading and writing to a removable non-volatile optical disc (e.g., a compact disc read-only memory (CD-ROM), a digital video disc read-only memory (DVD-ROM), or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this disclosure.

[0169] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of this disclosure.

[0170] Electronic device 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable human interaction with electronic device 12, and / or with any device that enables electronic device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, electronic device 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of electronic device 12 via bus 18. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0171] The processing unit 16 executes various functional applications and data processing by running programs stored in the memory 28, such as implementing the information recommendation method mentioned in the foregoing embodiments.

[0172] To implement the above embodiments, this disclosure also proposes an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided in the foregoing embodiments.

[0173] To implement the above embodiments, this disclosure also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

[0174] To implement the above embodiments, this disclosure also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

[0175] To implement the above embodiments, this disclosure also proposes a processor for invoking computer execution instructions to perform the methods provided in the foregoing embodiments.

[0176] In some embodiments of this disclosure, the processor may be, for example, a CPU, an ISP, etc., and there is no limitation thereto.

[0177] Those skilled in the art will also understand that the various illustrative logical blocks and steps listed in the embodiments of this disclosure can be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented in hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the described functionality using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of this disclosure.

[0178] Furthermore, the term “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as advantageous compared to other aspects or designs. Rather, the use of the term “exemplary” is intended to present the concept in a concrete manner. As used herein, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from the context, “X applies A or B” is intended to mean any of the natural inclusive arrangements. That is, “X applies A or B” satisfies any of the foregoing instances if X applies A; X applies B; or both X applies A and B. Additionally, unless otherwise specified or clear from the context to refer to the singular form, the articles “a” and “an” as used in this application and the appended claims are generally understood to mean “one or more.”

[0179] Similarly, although this disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding the specification and drawings. This disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components described above (e.g., elements, resources, etc.), unless otherwise indicated, the terminology used to describe such components is intended to correspond to any component (functionally equivalent) that performs the specific function of the described component, even if structurally not equivalent to the disclosed structure. Furthermore, although specific features of this disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations, as may be desired and advantageous to any given or particular application. Moreover, with regard to the terms “comprising,” “owning,” “having,” “having,” or variations thereof as used in the detailed description or claims, such terms are intended to be inclusive in a manner similar to the term “including.”

[0180] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0181] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. An information recommendation method, characterized in that, include: Obtain information about the first device and the device detection characteristics of the first device; Based on the first device information, first habit information is obtained, wherein the first habit information includes: at least one first conditional feature associated with the first device information, and first operation information associated with the first conditional feature; Determine a first condition feature that matches the device detection feature from at least one first condition feature, and use the matching first condition feature as the target condition feature; The first operation information associated with the target condition feature is used as the recommended target operation information.

2. The method according to claim 1, characterized in that, The first conditional feature includes: behavioral features and / or scene features, wherein the behavioral features include features in the object's behavioral dimension, and the scene features include features in the scene dimension; Wherein, determining a first condition feature that matches the device detection feature from at least one first condition feature, and using the matching first condition feature as a target condition feature, includes: Select a behavioral feature that matches the device detection feature from at least one of the behavioral features, wherein the device detection feature is used to describe object behavior; and / or, Scene features that match the device detection features are selected from at least one of the scene features, wherein the device detection features are used to describe the scene; wherein the matching behavioral features and / or the matching scene features are used as the target condition features.

3. The method according to claim 1, characterized in that, The step of obtaining the first habit information based on the first device information includes: Multiple candidate habit information is obtained, wherein the candidate habit information includes: candidate device information, candidate condition features associated with the candidate device information, and candidate operation information associated with the candidate condition features; Select candidate device information that is the same as the first device information from multiple candidate device information, and take the candidate habit information to which the candidate device information that is the same as the first device information belongs as the first habit information; Wherein, the candidate condition features in the candidate habit information are used as the first condition features, and the candidate operation information in the candidate habit information are used as the first operation information.

4. The method according to claim 3, characterized in that, The candidate habit information is pre-generated based on the following method: Multiple candidate behavior templates and multiple object behavior data are obtained, wherein the candidate behavior template includes: trigger feature, template condition feature corresponding to the trigger feature, and execution information corresponding to the template condition feature; The candidate habit information is generated based on the multiple candidate behavior templates and multiple object behavior data.

5. The method according to claim 4, characterized in that, The step of generating the candidate habit information based on the multiple candidate behavior templates and multiple object behavior data includes: Based on the multiple object behavior data and the candidate behavior template, a statistical reference value corresponding to each statistical dimension is determined, wherein the statistical reference value is used to quantitatively describe the degree of matching between the multiple object behavior data and the candidate behavior template under the statistical dimension; Based on the statistical reference value, select the target behavior template corresponding to the statistical dimension from multiple candidate behavior templates; The candidate habit information is generated based on the target behavior template.

6. The method according to claim 5, characterized in that, The step of determining the statistical reference value corresponding to each statistical dimension based on the multiple object behavior data and the candidate behavior template includes: Based on each of the statistical dimensions, when the current object behavior data matches the first part of the candidate behavior template, a preset value is added to the first value to obtain a first target value, wherein the first value is determined based on the previous object behavior data; Based on each of the statistical dimensions, when the current object behavior data and the second part of the candidate behavior template match, the second value is accumulated by the preset value to obtain the second target value, wherein the second value is determined based on the previous object behavior data; Based on the first target value, the second target value, and the statistical information corresponding to each statistical dimension, a statistical reference value corresponding to each statistical dimension is determined.

7. The method according to claim 6, characterized in that, The step of determining the statistical reference value corresponding to each statistical dimension based on the first target value, the second target value, and the statistical information corresponding to each statistical dimension includes: Based on the first target value and the second target value, determine the first descriptive value corresponding to each statistical dimension; Based on the second target value and the statistical information corresponding to each statistical dimension, determine the second descriptive value corresponding to each statistical dimension; Based on the first descriptive value and the second descriptive value, determine the statistical reference value corresponding to each statistical dimension.

8. The method according to claim 6, characterized in that, in, The first part includes: the triggering feature, the template condition feature, and the execution information; The second part includes: the triggering feature and the template condition feature.

9. The method according to claim 5, characterized in that, The step of selecting the target behavior template corresponding to the statistical dimension from multiple candidate behavior templates based on the statistical reference value includes: Select a statistical reference value that is greater than a threshold from a plurality of statistical reference values, and use the candidate behavior template corresponding to the selected statistical reference value as the target behavior template.

10. The method according to claim 5, characterized in that, The step of generating the candidate habit information based on the target behavior template includes: The candidate condition features are generated based on the template condition features in the target behavior template and at least some object behavior data; The candidate device information is determined based on the triggering features in the target behavior template; The candidate operation information is generated based on the execution information in the target behavior template and at least some object behavior data; The candidate habit information is generated based on the candidate condition features, the candidate device information, and the candidate operation information.

11. An information recommendation device, characterized in that, include: The first acquisition module is used to acquire first device information and device detection features of the first device; The second acquisition module is used to acquire first habit information based on the first device information, wherein the first habit information includes: at least one first conditional feature associated with the first device information, and first operation information associated with the first conditional feature; A determining module is configured to determine a first condition feature that matches the device detection feature from at least one first condition feature, and to use the matching first condition feature as a target condition feature; The recommendation module is used to use the first operation information associated with the target condition feature as the recommended target operation information.

12. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-10.

14. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1-10.

Images