A user portrait generation method and device, and an electronic device

Abstract

The application discloses a user portrait generation method and device and electronic equipment. The method comprises the following steps: in response to detecting a plurality of current viewing sequences, obtaining a to-be-detected device identifier and a current feature vector corresponding to each current viewing sequence; performing clustering operation on the plurality of current feature vectors to obtain at least one current clustering cluster; determining a current user portrait corresponding to each current viewing sequence based on a current mapping list corresponding to the plurality of current viewing sequences and at least one to-be-detected device identifier corresponding to each current clustering cluster; wherein the current mapping list contains at least two user portraits and a plurality of last device identifiers corresponding to each user portrait, and the current mapping list is determined based on last detected last feature vectors corresponding to a plurality of last viewing sequences. The embodiment of the application solves the problem that the existing viewing mapping list depends on human experience, and improves the prediction accuracy of the user portrait.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a method, device and electronic device for generating a user portrait. Background Art

[0002] User portraits usually refer to the extraction of data labels of different dimensions based on some user behaviors on the Internet. User portraits can help big data "get out" of the data warehouse. Diverse services such as personalized recommendations, precision marketing and advertising for users are an important direction for the implementation of big data.

[0003] Over-the-top (OTT) refers to internet companies developing various video and data services over the open internet, bypassing carriers. Currently, the most common method for capturing user profiles in the OTT industry is to manually pre-set a viewing mapping list. This list contains different viewing content and the corresponding user profiles. When a viewing content is obtained, the target user profile corresponding to that content is determined by querying the viewing mapping list.

[0004] In the existing technology, the setting of the movie viewing mapping list is greatly influenced by human experience, and people's movie viewing content may migrate over time, but the movie viewing mapping list is fixed and cannot be effectively updated, resulting in low accuracy of the predicted user portrait. Summary of the Invention

[0005] The present invention provides a method, device and electronic device for generating a user portrait, so as to solve the problem that the existing movie viewing mapping list relies on human experience and improve the prediction accuracy of the user portrait.

[0006] According to one aspect of the present invention, a method for generating a user portrait is provided, the method comprising:

[0007] In response to detecting a plurality of current movie viewing sequences, obtaining a to-be-detected device identifier and a current feature vector corresponding to each of the current movie viewing sequences;

[0008] Performing a clustering operation on the multiple current feature vectors to obtain at least one current cluster;

[0009] Determining, based on the current mapping lists corresponding to the multiple current movie viewing sequences and the at least one to-be-detected device identifier corresponding to each of the current clusters, the current user portrait corresponding to each of the current movie viewing sequences;

[0010] Among them, the current mapping list contains at least two user portraits and multiple previous device identifiers corresponding to each of the user portraits. The current mapping list is determined based on the previous feature vectors corresponding to multiple previous movie viewing sequences detected last time.

[0011] According to another aspect of the present invention, a device for generating a user portrait is provided, the device comprising:

[0012] a current feature vector acquisition module, configured to, in response to detecting a plurality of current movie viewing sequences, acquire a to-be-detected device identifier and a current feature vector corresponding to each of the current movie viewing sequences;

[0013] A current cluster determination module is used to perform a clustering operation on multiple current feature vectors to obtain at least one current cluster;

[0014] a current user portrait determination module, configured to determine a current user portrait corresponding to each current movie viewing sequence based on a current mapping list corresponding to the multiple current movie viewing sequences and at least one to-be-detected device identifier corresponding to each current cluster;

[0015] Among them, the current mapping list contains at least two user portraits and multiple previous device identifiers corresponding to each of the user portraits. The current mapping list is determined based on the previous feature vectors corresponding to multiple previous movie viewing sequences detected last time.

[0016] According to another aspect of the present invention, an electronic device is provided, comprising:

[0017] at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for generating a user portrait described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for generating a user portrait described in any embodiment of the present invention when executed.

[0021] The technical solution of the embodiment of the present invention obtains the device identifiers and current feature vectors to be detected corresponding to multiple current movie viewing sequences detected, performs a clustering operation on the multiple current feature vectors, obtains at least one current clustering cluster, and determines the current user portraits corresponding to each current movie viewing sequence based on the current mapping lists corresponding to the multiple current movie viewing sequences and the at least one device identifier to be detected corresponding to each current clustering cluster. The current mapping list is determined based on the previous feature vectors corresponding to the multiple previous movie viewing sequences detected last time, which solves the problem of the existing movie viewing mapping list relying on human experience settings, makes the current mapping list have time migration, and thereby improves the prediction accuracy of user portraits.

[0022] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 This is a flowchart of a method for generating a user portrait provided in Example 1 of the present invention;

[0025] Figure 2 is a schematic diagram of a specific example of a method for generating a user portrait provided by an embodiment of the present invention;

[0026] Figure 3 This is a flowchart of a method for generating a user portrait provided in the second embodiment of the present invention;

[0027] Figure 4 This is a flowchart of a method for generating a user portrait provided by the third embodiment of the present invention;

[0028] Figure 5 This is a schematic diagram of the structure of a device for generating a user portrait provided by a fourth embodiment of the present invention;

[0029] Figure 6 This is a structural diagram of an electronic device provided in Example 5 of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] Example 1

[0033] Figure 1 This is a flow chart of a method for generating a user portrait provided by the first embodiment of the present invention. This embodiment is applicable to the case of predicting a user portrait based on a movie viewing sequence. The method can be executed by a user portrait generating device, which can be implemented in the form of hardware and / or software. The user portrait generating device can be configured in a terminal device. Figure 1 As shown, the method includes:

[0034] S110 . In response to detecting multiple current movie viewing sequences, obtaining a to-be-detected device identifier and a current feature vector corresponding to each current movie viewing sequence.

[0035] Specifically, the current viewing sequence can be used to characterize the viewing information data corresponding to at least one viewing work from the start-up moment to the end of the shutdown moment, wherein the power-on moment to the shutdown moment constitutes the viewing time corresponding to the current viewing sequence. Exemplarily, the viewing information data includes but is not limited to the name of the TV series, the type of TV series, the release time, the name of the main actor of the TV series, the name of the director of the TV series, the length of a single episode of the TV series, and the introduction of the TV series, etc. For example, when the viewing work includes the TV series AA, the viewing information data corresponding to the viewing work include "AA", "Romance & Costume", "2022", "Name of the male protagonist A", "Name of the female protagonist B", "Name of the director C", "45 minutes" and "Love, hate and hatred in ancient times", etc.

[0036] In one embodiment, the method optionally further includes: determining a plurality of filtered current viewing sequences based on the viewing durations corresponding to the plurality of current viewing sequences and a preset duration threshold. Specifically, if the viewing duration of current viewing sequence A is greater than or equal to the preset duration threshold, current viewing sequence A is used as the filtered current viewing sequence; if the viewing duration of current viewing sequence A is less than the preset duration threshold, current viewing sequence A is deleted. Exemplarily, the preset duration threshold may be 10 minutes.

[0037] The advantage of this setting is that the coupling between the current viewing sequence with a shorter viewing time and the user portrait is relatively low. Since the embodiment of the present invention determines the user portrait in a clustering manner, if the current viewing sequence with a shorter viewing time is added to the data participating in the clustering, it is easy to affect the overall clustering effect, resulting in inaccurate prediction results of the user portrait.

[0038] Specifically, the device identifier to be detected can be used to represent the device identifier corresponding to the terminal device that has collected the current viewing sequence, and the same device identifier to be detected can correspond to at least one current viewing sequence. Exemplarily, the device identifier to be detected can be the MAC (Media Access Control Address) device identifier of the terminal device. The coded content of the device identifier to be detected includes at least one of text, numbers, letters and special characters. The coded content of the device identifier to be detected is not limited here, and users can customize it according to actual needs.

[0039] In one embodiment, optionally, obtaining current feature vectors corresponding to each current viewing sequence includes: for each current viewing sequence, inputting viewing information data corresponding to at least one viewing work corresponding to the current viewing sequence into a pre-trained model, obtaining output initial feature vectors corresponding to each viewing work, and performing an average operation on each initial feature vector to obtain the current feature vector corresponding to the current viewing sequence. The pre-trained model has the ability to extract semantic information. Exemplary pre-trained models include, but are not limited to, a BERT model, an ELMo model, or a GPT model.

[0040] S120: Perform a clustering operation on multiple current feature vectors to obtain at least one current cluster.

[0041] Exemplarily, the clustering algorithm used in the clustering operation includes, but is not limited to, a k-means clustering algorithm, a mean shift clustering algorithm, a density-based clustering algorithm, a k-medoids clustering algorithm, or a clarans clustering algorithm, etc. The clustering algorithm used is not limited here.

[0042] Specifically, each current cluster includes at least one current feature vector, and each current feature vector corresponds to a current movie viewing sequence and an identifier of a device to be detected.

[0043] S130 : Determine the current user portrait corresponding to each current movie viewing sequence based on the current mapping lists corresponding to the multiple current movie viewing sequences and at least one to-be-detected device identifier corresponding to each current cluster.

[0044] Among them, illustratively, the current user portrait includes but is not limited to at least one of the user's age group, user gender and user identity attributes, wherein the user identity attribute can be a literary youth, an expert scholar, a white-collar worker or an uneducated person, etc.

[0045] In this embodiment, the current mapping list includes at least two user portraits and multiple previous device identifiers corresponding to each user portrait. The current mapping list is determined based on the previous feature vectors corresponding to the multiple previous movie viewing sequences detected last time. For example, when the user portrait only includes the user's age group, the current mapping list may include 0-17 years old, 18-24 years old, 25-34 years old, 35-44 years old, 45-54 years old, and over 55 years old. When the user portrait includes the user's age group and gender, the current mapping list may include 0-17 years old female, 0-17 years old male, 18-24 years old female, 18-24 years old male, and so on.

[0046] In one embodiment, optionally, the method further includes: obtaining at least two previous clusters corresponding to multiple previous movie viewing sequences, user portraits corresponding to each previous cluster, and a previous device identification set; wherein each previous cluster is obtained based on the clustering of previous feature vectors corresponding to each previous movie viewing sequence, and the previous device identification set includes the previous device identifications of multiple previous movie viewing sequences corresponding to the previous cluster; for each previous cluster, obtaining any previous device identification in the previous device identification set corresponding to the previous cluster, and if there is no intersection between the previous device identification and other previous device identification sets, adding the previous device identification and the user portrait corresponding to the previous cluster to the current mapping list.

[0047] Specifically, the previous viewing sequence is the viewing sequence detected last time, wherein the previous detection time of the previous viewing sequence and the current detection time of the current viewing sequence meet a preset detection period. Exemplarily, the preset detection period can be 1 day or one week.

[0048] Specifically, if each previous viewing sequence is the first detected viewing sequence, the user profile corresponding to each previous cluster can be set by an expert based on the viewing sequence corresponding to each previous cluster. If each previous viewing sequence is not the first detected viewing sequence, the user profile corresponding to each previous cluster can be determined based on the previous mapping list corresponding to each previous viewing sequence.

[0049] Specifically, each previous device identifier in the current mapping list can uniquely identify each user portrait, that is, the previous device identifier corresponding to the user portrait does not exist in the previous cluster corresponding to another user portrait. The number of samples of the previous device identifiers corresponding to the user portraits in the current mapping list obtained by sampling can be determined according to the size of each previous cluster. For example, assuming that the previous device identifier set A corresponding to the previous cluster A contains 10 previous device identifiers, and the previous device identifier set B corresponding to the previous cluster B contains 5 previous device identifiers, then the number of samples is less than 5.

[0050] In one embodiment, optionally, based on the current mapping lists corresponding to multiple current movie viewing sequences and at least one device identifier to be detected corresponding to each current clustering cluster, the current user portraits corresponding to each current movie viewing sequence are determined, including: for each current clustering cluster, based on at least one device identifier to be detected corresponding to the current clustering cluster, determining the number of device identifier overlaps corresponding to at least two user portraits in the current mapping list; and using the user portrait with the largest number of device identifier overlaps as the current user portrait of each current movie viewing sequence corresponding to the current clustering cluster.

[0051] Among them, for example, assuming that the current cluster A corresponds to 10 device identifiers to be detected, namely 1-10, and the previous device identifier corresponding to user portrait A in the current mapping list includes [1-4, 11-20], and the previous device identifier corresponding to user portrait B includes [5-10, 21-31], then the number of overlapping device identifiers corresponding to user portrait A is 4, and the number of overlapping device identifiers corresponding to user portrait B is 6, then the target user portrait of each current viewing sequence corresponding to the current cluster A is user portrait B.

[0052] Figure 2 is a schematic diagram of a specific example of a method for generating a user portrait provided by an embodiment of the present invention. Figure 2Take the preset detection period of 1 day as an example. Specifically, determine the first feature vector corresponding to the first viewing sequence detected on the 1st day, perform clustering operations on each first feature vector, and obtain two first clusters. Experts can set user portraits corresponding to each first cluster according to the first viewing sequence corresponding to each first cluster. And sample the first device identifier corresponding to each first cluster to obtain mapping list 1. Determine the second feature vector corresponding to the second viewing sequence detected on the 2nd day, perform clustering operations on each second feature vector, and obtain two second clusters. Based on mapping list 1 and the second device identifier corresponding to each second cluster, determine the user portrait of the viewing sequence corresponding to each second cluster. And sample the second device identifier corresponding to each second cluster to obtain mapping list 2. Similarly, determine the Nth feature vector corresponding to the Nth movie viewing sequence detected on the Nth day, perform a clustering operation on each Nth feature vector to obtain two Nth clusters, and determine the user profile of the movie viewing sequence corresponding to each Nth cluster based on the mapping list N-1 and the Nth device identifier corresponding to each Nth cluster.

[0053] The advantage of this setup is that the existing mapping list requires adding the corresponding user profile for each new film or TV show released. However, films and TV shows are constantly being released and updated very quickly. Using clustering, we can find similar clusters based on the feature vectors of new films and TV shows, effectively compensating for the need to manually update the mapping list dynamically and ensuring the accuracy of user profile prediction.

[0054] The technical solution of this embodiment obtains the device identifiers and current feature vectors to be detected corresponding to multiple current movie viewing sequences, performs a clustering operation on the multiple current feature vectors, obtains at least one current clustering cluster, and determines the current user portraits corresponding to each current movie viewing sequence based on the current mapping lists corresponding to the multiple current movie viewing sequences and the at least one device identifier to be detected corresponding to each current clustering cluster. The current mapping list is determined based on the previous feature vectors corresponding to the multiple previous movie viewing sequences detected last time, which solves the problem of the existing movie viewing mapping list relying on human experience, makes the current mapping list have time migration, and thereby improves the prediction accuracy of the user portrait.

[0055] Terminal devices in the OTT industry have common attributes, that is, one terminal device usually corresponds to all members of a family. How to split the viewing behavior corresponding to a terminal device into the perspective of each family member, that is, to distinguish the viewing behavior of a terminal device and build a user portrait corresponding to each member of the family, is of great significance for personalized recommendations and refined positioning of precision marketing.

[0056] Example 2

[0057] Figure 3 This is a flow chart of a method for generating a user portrait provided by the second embodiment of the present invention. This embodiment further refines the method of determining the different viewing time habits of family members of a family to which a single terminal device belongs after predicting the user portrait in the above embodiment. Figure 3 As shown, the method includes:

[0058] S210 : In response to detecting multiple current movie viewing sequences, obtaining the to-be-detected device identifier and the current feature vector corresponding to each current movie viewing sequence.

[0059] S220 , performing a clustering operation on multiple current feature vectors to obtain at least one current cluster.

[0060] S230 : Determine the current user portrait corresponding to each current movie viewing sequence based on the current mapping lists corresponding to the multiple current movie viewing sequences and at least one to-be-detected device identifier corresponding to each current cluster.

[0061] S240. For each device identifier to be detected, obtain historical startup times and historical user portraits corresponding to multiple historical movie viewing sequences corresponding to the device identifier to be detected.

[0062] In this embodiment, the historical viewing sequence includes multiple current viewing sequences and / or previous viewing sequences corresponding to the device identifier to be detected. Specifically, the same device identifier to be detected usually corresponds to multiple viewing sequences. Based on the preset statistical period, the viewing sequence corresponding to the device identifier to be detected that has been detected in the past is used as the historical viewing sequence corresponding to the device identifier to be detected. For example, assuming that the preset statistical period is 10 days and the preset detection period is 1 day, the historical viewing sequence includes the viewing sequences corresponding to the device identifier to be detected detected from the i-th day to the i+9-th day.

[0063] For example, assuming that the preset statistical period is 2 days and the preset detection period is 1 day, viewing sequence 1 and viewing sequence 2 are detected on the i-th day, and their corresponding device identifiers to be detected are device identifier A and device identifier B respectively, viewing sequence 3 and viewing sequence 4 are detected on the i+1-th day, and their corresponding device identifiers to be detected are device identifier A and device identifier C respectively, then for device identifier A, the multiple historical viewing sequences corresponding to it include viewing sequence 1 and viewing sequence 3.

[0064] S250: Determine a time portrait mapping list based on each historical boot time and each historical user portrait.

[0065] In this embodiment, the time profile mapping list includes at least two preset time periods and the classified user profiles corresponding to each preset time period. Specifically, the time profile mapping list can be used to represent information about the different viewing time habits of family members of the family to which the terminal device corresponding to the device to be detected belongs.

[0066] Specifically, the time intervals corresponding to each preset time period may be the same or different. In one embodiment, the preset time period may be a time period corresponding to 24 hours in a day, such as multiple preset time periods are 0-7 o'clock, 7-12 o'clock, 12-18 o'clock, and 18-24 o'clock. In another embodiment, the preset time period may be a time period corresponding to 7 days in a week, such as multiple preset time periods are Monday to Friday, Saturday, and Sunday. In another embodiment, the preset time period may also be a time period corresponding to 7 days in a week and 24 hours in a day, such as multiple preset time periods are 0-7 o'clock from Monday to Friday, 7-12 o'clock from Monday to Friday, 12-18 o'clock from Monday to Friday, 18-24 o'clock from Monday to Friday, 0-7 o'clock on Saturday, 7-24 o'clock on Saturday, and 0-24 o'clock on Sunday. The specific parameter values ​​of each preset time period are not limited here, and users can customize the settings according to actual needs.

[0067] The following contents of this embodiment are all described illustratively using 24 hours in one day as an example.

[0068] For example, the time profile mapping list includes 0-7 am, 7-12 pm, 12-18 pm, and 6-12 am, and the classified user profiles corresponding to each preset time period are User Profile A, User Profile B, User Profile A, and User Profile C. It is understandable that the user profiles corresponding to different preset time periods can be the same or different. For example, when the terminal device is a TV, the elderly generally watch TV from 7-12 am and from 6-12 am, while young people generally only watch TV from 6-12 am.

[0069] Specifically, for each preset time period, based on each historical power-on time, multiple target historical movie viewing sequences corresponding to the preset time period are determined, the number of sequences of each target historical movie viewing sequence corresponding to at least one user portrait is determined, and the user portrait with the largest number of sequences is used as the classified user portrait corresponding to the preset time period.

[0070] For example, assuming that the multiple historical viewing sequences corresponding to the device identification to be detected include historical viewing sequence 1, historical viewing sequence 2, historical viewing sequence 3 and historical viewing sequence 4, and the corresponding historical power-on times are 8:00, 8:03, 9:02 and 20:23 respectively, assuming that the preset time period is 7-12 o'clock, the multiple target historical viewing sequences corresponding to 7-12 o'clock include historical viewing sequence 1, historical viewing sequence 2 and historical viewing sequence 3, assuming that the user portraits corresponding to historical viewing sequence 1, historical viewing sequence 2 and historical viewing sequence 3 are user portrait A, user portrait A and user portrait B respectively, then the number of sequences corresponding to user portrait A is 2, the number of sequences corresponding to user portrait B is 1, and user portrait A with the largest number of sequences is used as the classified user portrait corresponding to 7-12 o'clock.

[0071] Based on the above embodiment, optionally, the method also includes: in response to detecting a power-on instruction corresponding to the target device to be detected identifier, determining the target classification user portrait corresponding to the power-on instruction based on the target time portrait mapping list corresponding to the target device to be detected identifier and the current power-on time corresponding to the power-on instruction; obtaining recommendation data corresponding to the target classification user portrait, and sending the recommendation data to the terminal device corresponding to the target device to be detected identifier, so that the terminal device displays the recommended data.

[0072] Specifically, the local storage space stores multiple identifications of devices to be detected and a time portrait mapping list corresponding to each identification of the devices to be detected.

[0073] Among them, for example, it is assumed that the target time portrait mapping list corresponding to the target device identification to be detected includes 0-7 o'clock, 7-12 o'clock, 12-18 o'clock and 18-24 o'clock, and the classified user portraits corresponding to each preset time period are user portrait A, user portrait B, user portrait A and user portrait C respectively. Assuming that the current power-on time is 12:05, the target classified user portrait is user portrait B, and the recommended data corresponding to user portrait B is obtained.

[0074] Exemplarily, recommended data includes, but is not limited to, shopping data, advertising data, recommended film and television data, etc. For example, when the target user profile is children, the recommended data may be advertising data related to children's snacks or children's toys; when the target user profile is the elderly, the recommended data may be advertising data related to health products.

[0075] The technical solution of this embodiment obtains the historical power-on time and historical user portraits corresponding to multiple historical viewing sequences corresponding to each device identification to be detected, and determines a time portrait mapping list based on each historical power-on time and each historical user portrait, thereby solving the problem of being unable to accurately distinguish the viewing time habits of family members of the family to which the terminal device belongs. The viewing behavior of each family member of the family to which a certain terminal device belongs can be precisely positioned, achieving the purpose of predicting family portraits and providing important data reference for subsequent recommendations or marketing services.

[0076] Example 3

[0077] Figure 4 This is a flow chart of a method for generating a user portrait provided by the third embodiment of the present invention. This embodiment further refines the method of determining the family structure of the family to which a single terminal device belongs after predicting the user portrait in the above embodiment. Figure 4 As shown, the method includes:

[0078] S310 : In response to detecting multiple current movie viewing sequences, obtaining the to-be-detected device identifier and the current feature vector corresponding to each current movie viewing sequence.

[0079] S320: Perform a clustering operation on multiple current feature vectors to obtain at least one current cluster.

[0080] S330 : Determine the current user portrait corresponding to each current movie viewing sequence based on the current mapping lists corresponding to the multiple current movie viewing sequences and at least one device identifier to be detected corresponding to each current cluster.

[0081] S340. For each device identifier to be detected, obtain historical user portraits corresponding to multiple historical movie viewing sequences corresponding to the device identifier to be detected.

[0082] In this embodiment, the historical viewing sequence includes multiple current viewing sequences and / or previous viewing sequences corresponding to the device identifier to be detected. Specifically, the same device identifier to be detected usually corresponds to multiple viewing sequences. Based on the preset statistical period, the viewing sequence corresponding to the device identifier to be detected that has been detected in the past is used as the historical viewing sequence corresponding to the device identifier to be detected. For example, assuming that the preset statistical period is 10 days and the preset detection period is 1 day, the historical viewing sequence includes the viewing sequences corresponding to the device identifier to be detected detected from the i-th day to the i+9-th day.

[0083] For example, assuming that the preset statistical period is 2 days and the preset detection period is 1 day, viewing sequence 1 and viewing sequence 2 are detected on the i-th day, and their corresponding device identifiers to be detected are device identifier A and device identifier B respectively, viewing sequence 3 and viewing sequence 4 are detected on the i+1-th day, and their corresponding device identifiers to be detected are device identifier A and device identifier C respectively, then for device identifier A, the multiple historical viewing sequences corresponding to it include viewing sequence 1 and viewing sequence 3.

[0084] In this embodiment, the historical user profile includes the user's age group or the historical user profile includes the user's age group and the user's gender. In one embodiment, the historical user profile includes, for example, 0-17 years old or 18-24 years old. In another embodiment, the historical user profile includes, for example, female 0-17 years old, male 0-17 years old, or female 18-24 years old.

[0085] The following content of this embodiment is exemplified by taking the historical user profile including the user age group.

[0086] S350: Determine the number of historical movie viewing sequences corresponding to at least one historical user portrait, and based on the number of each sequence, determine the family structure corresponding to the device identifier to be detected.

[0087] Among them, for example, assuming that the multiple historical movie viewing sequences corresponding to the device identification to be detected include historical movie viewing sequence 1, historical movie viewing sequence 2, historical movie viewing sequence 3 and historical movie viewing sequence 4, and their corresponding historical user portraits are 0-17 years old, 0-17 years old, 18-24 years old and 25-34 years old respectively, then the number of sequences corresponding to 0-17 years old is 2, and the number of sequences corresponding to 18-24 years old and 25-34 years old are 1 respectively.

[0088] In one embodiment, optionally, based on the number of each sequence, the family structure corresponding to the identification of the device to be detected is determined, including: based on preset selection parameters and the number of each sequence, determining at least one target historical user portrait, and determining the family structure corresponding to the identification of the device to be detected based on each target historical user portrait; wherein the preset selection parameters include a preset selection number or a preset number threshold; and / or, for the number of sequences corresponding to each historical user portrait, if the historical user portrait corresponding to the current sequence number belongs to the middle age group, then when the current sequence number is greater than or equal to the next sequence number and greater than the previous sequence number, the historical user portrait corresponding to the current sequence number is used as the target historical user portrait; if the historical user portrait corresponding to the current sequence number belongs to the age groups at both ends, then when the current sequence number is greater than or equal to the next sequence number or greater than the previous sequence number, the historical user portrait corresponding to the current sequence number is used as the target historical user portrait; based on each target historical user portrait, the family structure corresponding to the identification of the device to be detected is determined.

[0089] In one embodiment, specifically, when the preset selection parameter is a preset selection number, the number of sequences is sorted in descending order, and the historical user profiles with the preset selection number at the top of the sorting results are used as target historical user profiles. When the preset selection parameter is a preset number threshold, the historical user profiles with a sequence number greater than the preset number threshold are used as target historical user profiles.

[0090] For example, assuming that the number of sequences corresponding to 0-17 years old, 18-24 years old, 25-34 years old, 35-44 years old, 45-54 years old, and over 55 years old are 100, 2, 50, 7, 10, and 1 respectively, if the preset selection number is 3, then at least one target historical user profile includes 0-17 years old, 25-34 years old, and 45-54 years old. If the preset selection threshold is 40, then at least one target historical user profile includes 0-17 years old and 25-34 years old. Accordingly, when at least one target historical user profile includes 0-17 years old, 25-34 years old, and 45-54 years old, the family structure is three generations living together. When at least one target historical user profile includes 0-17 years old and 25-34 years old, the family structure is two generations living together.

[0091] In another embodiment, specifically, the middle age segment is used to characterize that the historical user profile has both the previous historical user profile adjacent to it and the next age segment adjacent to it, and the end age segments are used to characterize that the historical user profile only has the previous historical user profile adjacent to it or only has the next age segment adjacent to it. For example, assuming that multiple historical user profiles include 0-17 years old, 18-24 years old, 25-34 years old, 35-44 years old, 45-54 years old and over 55 years old, then "0-17 years old" and "over 55 years old" both belong to the end age segments, and 8-24 years old, 25-34 years old, 35-44 years old, and 45-54 years old all belong to the middle age segment.

[0092] Specifically, when the historical user profile belongs to the middle age group, the number of this sequence is greater than the number of the previous sequence, and the number of this sequence is greater than or equal to the number of the next sequence. This is because viewing sequences are generally shifted backward. That is, over time, the same user tends to watch viewing sequences corresponding to older historical user profiles. For example, if user A initially watches children's animation, over time, user A may begin to watch youth animation.

[0093] For example, assuming that the number of sequences corresponding to 0-17 years old, 18-24 years old, 25-34 years old, 35-44 years old, 45-54 years old, and over 55 years old are 100, 2, 50, 7, 10, and 1 respectively, then at least one target historical user profile includes 0-17 years old, 25-34 years old, and 45-54 years old, and accordingly, the family structure is three generations living under one roof.

[0094] In another embodiment, optionally, when a family structure is obtained respectively by the above three methods, it is determined whether the three family structures are the same. If so, any family structure is used as the family structure corresponding to the identification of the device to be detected. If not, the family structure corresponding to the identification of the device to be detected is set to pending or the family structure determined based on the method of comparing the number of adjacent sequences is used as the family structure corresponding to the identification of the device to be detected.

[0095] Methods based on comparing a preset number threshold or sorting the number of sequences require setting the parameter values ​​for the preset selection number and the preset number threshold in advance based on manual experience. These parameter values ​​directly affect the number of target historical user profiles determined. For example, if the preset selection number is three, the number of target historical user profiles must be three. The method based on comparing the number of adjacent sequences effectively avoids the problem of relying on manual experience, and the family structure determined is slightly more accurate than the family structures determined by the above two methods. Therefore, if the three family structures are different, the family structure determined by the method based on comparing the number of adjacent sequences can optionally be used as the family structure corresponding to the device identification to be detected.

[0096] In another embodiment, when the historical user portrait includes the user age group and the user gender, the family structure determined based on at least one target historical user portrait is more detailed than the family structure determined based on at least one target user age group. Exemplarily, when the historical user portrait includes the user age group, assuming that at least one target historical user portrait includes 0-17 years old and 25-34 years old, the family structure is two generations living together. When the historical user portrait includes the user age group and the user gender, assuming that at least one target historical user portrait includes female 0-17 years old, female 25-34 years old and male 25-34 years old, the family structure is a family of three, and assuming that at least one target historical user portrait includes female 0-17 years old and female 25-34 years old, the family structure is a female single-parent family.

[0097] Based on the above embodiment, optionally, in response to detecting a power-on instruction corresponding to the target device to be detected identifier, recommended data is obtained based on the target family structure corresponding to the target device to be detected identifier; and the recommended data is sent to a terminal device corresponding to the target device to be detected identifier, so that the terminal device displays the recommended data.

[0098] Exemplarily, the recommended data includes, but is not limited to, shopping data, advertising data, recommended film and television data, etc. For example, when the target family structure is a three-generation household, the recommended data may be advertising data related to health products; when the target family structure is a single-parent family, the recommended data may be advertising data related to childcare centers.

[0099] The technical solution of this embodiment obtains, for each device identification to be detected, historical user portraits corresponding to multiple historical movie viewing sequences corresponding to the device identification to be detected, wherein the historical user portraits include the user's age group or the historical user portraits include the user's age group and the user's gender, determines the number of sequences of historical movie viewing sequences corresponding to at least one historical user portrait, and determines the family structure corresponding to the device identification to be detected based on the number of sequences, thereby solving the problem of being unable to accurately identify the family structure of the family to which the terminal device belongs, achieving the purpose of predicting the family structure, and providing important data reference for subsequent recommendation or marketing services.

[0100] Example 4

[0101] Figure 5 This is a schematic diagram of the structure of a device for generating a user portrait provided by the fourth embodiment of the present invention. Figure 5 As shown, the device includes: a current feature vector acquisition module 410, a current cluster determination module 420 and a current user portrait determination module 430.

[0102] The current feature vector acquisition module 410 is configured to acquire the device identifier to be detected and the current feature vector corresponding to each current viewing sequence in response to detecting multiple current viewing sequences;

[0103] A current cluster determination module 420 is configured to perform a clustering operation on a plurality of current feature vectors to obtain at least one current cluster;

[0104] A current user profile determination module 430 is configured to determine a current user profile corresponding to each current movie viewing sequence based on a current mapping list corresponding to multiple current movie viewing sequences and at least one to-be-detected device identifier corresponding to each current cluster;

[0105] Among them, the current mapping list contains at least two user portraits and multiple previous device identifiers corresponding to each user portrait. The current mapping list is determined based on the previous feature vectors corresponding to multiple previous movie viewing sequences detected last time.

[0106] The technical solution of this embodiment obtains the device identifiers and current feature vectors to be detected corresponding to multiple current movie viewing sequences, performs a clustering operation on the multiple current feature vectors, obtains at least one current clustering cluster, and determines the current user portraits corresponding to each current movie viewing sequence based on the current mapping lists corresponding to the multiple current movie viewing sequences and the at least one device identifier to be detected corresponding to each current clustering cluster. The current mapping list is determined based on the previous feature vectors corresponding to the multiple previous movie viewing sequences detected last time, which solves the problem of the existing movie viewing mapping list relying on human experience, makes the current mapping list have time migration, and thereby improves the prediction accuracy of the user portrait.

[0107] Based on the above embodiment, optionally, the device further includes:

[0108] A current mapping list determination module is configured to obtain at least two previous clusters corresponding to multiple previous viewing sequences, user profiles corresponding to each previous cluster, and a previous device identification set; wherein each previous cluster is obtained based on clustering of previous feature vectors corresponding to each previous viewing sequence, and the previous device identification set includes the previous device identifications of multiple previous viewing sequences corresponding to the previous cluster;

[0109] For each previous cluster, obtain any previous device identifier in the previous device identifier set corresponding to the previous cluster. If the previous device identifier does not have an intersection with other previous device identifier sets, add the previous device identifier and the user portrait corresponding to the previous cluster to the current mapping list.

[0110] Based on the above embodiment, optionally, the current user portrait determination module 430 is specifically configured to:

[0111] For each current cluster, based on at least one to-be-detected device identifier corresponding to the current cluster, determine the number of overlaps between the device identifiers corresponding to at least two user profiles in the current mapping list;

[0112] The user profile with the largest number of overlapping device identifiers is used as the current user profile of each current movie viewing sequence corresponding to the current cluster.

[0113] Based on the above embodiment, optionally, the device further includes:

[0114] A time profile mapping list determination module is configured to obtain, for each device identifier to be detected, the historical power-on time and historical user profile corresponding to each of the multiple historical viewing sequences corresponding to the device identifier to be detected; wherein the historical viewing sequences include multiple current viewing sequences and / or previous viewing sequences corresponding to the device identifier to be detected;

[0115] Based on each historical power-on time and each historical user portrait, a time portrait mapping list is determined; wherein the time portrait mapping list includes at least two preset time periods and classified user portraits corresponding to each preset time period.

[0116] Based on the above embodiment, optionally, the device further includes:

[0117] A first recommendation data sending module is configured to, in response to detecting a power-on instruction corresponding to a target device identifier to be detected, determine a target classification user profile corresponding to the power-on instruction based on a target time profile mapping list corresponding to the target device identifier to be detected and a current power-on time corresponding to the power-on instruction;

[0118] Obtain recommended data corresponding to the target classification user portrait, and send the recommended data to the terminal device corresponding to the target device identifier to be detected, so that the terminal device can display the recommended data.

[0119] Based on the above embodiment, optionally, the device further includes:

[0120] A family structure determination module is configured to obtain, for each device identifier to be detected, historical user profiles corresponding to multiple historical movie viewing sequences corresponding to the device identifier to be detected; wherein the historical movie viewing sequences include multiple current movie viewing sequences and / or previous movie viewing sequences corresponding to the device identifier to be detected, and the historical user profiles include user age groups or user age groups and user gender;

[0121] Determine the number of historical movie viewing sequences corresponding to at least one historical user profile, and determine the family structure corresponding to the device identification to be detected based on the number of each sequence.

[0122] Based on the above embodiment, the device may optionally further include: a family structure determination module, specifically configured to:

[0123] Based on preset selection parameters and the number of sequences, at least one target historical user profile is determined, and based on each target historical user profile, the family structure corresponding to the device identifier to be detected is determined; wherein the preset selection parameters include a preset selection number or a preset number threshold; and / or,

[0124] For each historical user profile, if the corresponding historical user profile belongs to the middle age group, then if the current sequence number is greater than or equal to the next sequence number and greater than the previous sequence number, the historical user profile corresponding to the current sequence number will be used as the target historical user profile.

[0125] If the historical user profile corresponding to the current sequence number belongs to the age groups at both ends, then when the current sequence number is greater than or equal to the next sequence number or greater than the previous sequence number, the historical user profile corresponding to the current sequence number will be used as the target historical user profile;

[0126] Based on the historical user portraits of each target, the family structure corresponding to the device identification to be detected is determined.

[0127] Based on the above embodiment, optionally, the device further includes:

[0128] a second recommended data sending module, configured to, in response to detecting a power-on instruction corresponding to the target device identifier to be detected, obtain recommended data based on a target family structure corresponding to the target device identifier to be detected;

[0129] The recommended data is sent to the terminal device corresponding to the target device to be detected, so that the terminal device displays the recommended data.

[0130] The user portrait generation device provided in the embodiment of the present invention can execute the user portrait generation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0131] Example 5

[0132] Figure 6 1 is a structural diagram of an electronic device provided in Example 5 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0133] like Figure 6As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0134] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0135] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the method for generating a user profile.

[0136] In some embodiments, the method for generating a user portrait may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for generating a user portrait described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the method for generating a user portrait in any other appropriate manner (e.g., by means of firmware).

[0137] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0138] Computer programs for implementing the user profile generation method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer programs can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0139] Example 6

[0140] Embodiment 6 of the present invention further provides a computer-readable storage medium storing computer instructions, which are used to cause a processor to execute a method for generating a user portrait, the method comprising:

[0141] In response to detecting multiple current movie viewing sequences, obtaining a to-be-detected device identifier and a current feature vector corresponding to each current movie viewing sequence;

[0142] Performing a clustering operation on the multiple current feature vectors to obtain at least one current cluster;

[0143] Determine, based on current mapping lists corresponding to multiple current movie viewing sequences and at least one to-be-detected device identifier corresponding to each current cluster, a current user profile corresponding to each current movie viewing sequence;

[0144] Among them, the current mapping list contains at least two user portraits and multiple previous device identifiers corresponding to each user portrait. The current mapping list is determined based on the previous feature vectors corresponding to multiple previous movie viewing sequences detected last time.

[0145] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0146] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0147] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0148] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0149] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0150] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for generating a user portrait, characterized in that: include: In response to detecting a plurality of current movie viewing sequences, obtaining a to-be-detected device identifier and a current feature vector corresponding to each of the current movie viewing sequences; Performing a clustering operation on the multiple current feature vectors to obtain at least one current cluster; Determining, based on the current mapping list and at least one to-be-detected device identifier corresponding to each of the current clusters, a current user portrait corresponding to each of the current movie viewing sequences; The current mapping list includes at least two user portraits and multiple previous device identifiers corresponding to each of the user portraits, and the current mapping list is determined based on the previous feature vectors corresponding to the multiple previous movie viewing sequences detected last time; The determining, based on the current mapping list and at least one to-be-detected device identifier corresponding to each current cluster, the current user portrait corresponding to each current movie viewing sequence includes: For each current cluster, based on at least one to-be-detected device identifier corresponding to the current cluster, determine the number of overlaps between the device identifiers corresponding to at least two user profiles in the current mapping list; The user portrait with the largest number of overlapping device identifications is used as the current user portrait of each current movie viewing sequence corresponding to the current cluster.

2. The method according to claim 1, characterized in that The method further comprises: Obtaining at least two previous clusters corresponding to multiple previous movie viewing sequences, user profiles corresponding to each of the previous clusters, and a previous device identification set; wherein each of the previous clusters is obtained based on clustering of previous feature vectors corresponding to each of the previous movie viewing sequences, and the previous device identification set includes the previous device identifications of the multiple previous movie viewing sequences corresponding to the previous clusters; For each previous cluster, obtain any previous device identifier in the previous device identifier set corresponding to the previous cluster. If there is no intersection between the previous device identifier and other previous device identifier sets, add the previous device identifier and the user portrait corresponding to the previous cluster to the current mapping list.

3. The method according to claim 1, characterized in that The method further comprises: For each device identifier to be detected, obtain historical boot time and historical user profiles corresponding to multiple historical movie viewing sequences corresponding to the device identifier to be detected; wherein the historical movie viewing sequences include multiple current movie viewing sequences and / or previous movie viewing sequences corresponding to the device identifier to be detected; Based on each of the historical power-on times and each of the historical user portraits, a time portrait mapping list is determined; wherein, the time portrait mapping list includes at least two preset time periods and classified user portraits corresponding to each of the preset time periods.

4. The method according to claim 3, characterized in that The method further comprises: In response to detecting a power-on instruction corresponding to the target device identifier to be detected, determining a target classification user profile corresponding to the power-on instruction based on a target time profile mapping list corresponding to the target device identifier to be detected and a current power-on time corresponding to the power-on instruction; Acquire recommendation data corresponding to the target classification user portrait, and send the recommendation data to a terminal device corresponding to the target device identifier to be detected, so that the terminal device displays the recommendation data.

5. The method according to claim 1, characterized in that The method further comprises: For each device identifier to be detected, obtain historical user portraits corresponding to multiple historical movie viewing sequences corresponding to the device identifier to be detected; wherein the historical movie viewing sequences include multiple current movie viewing sequences and / or previous movie viewing sequences corresponding to the device identifier to be detected, and the historical user portraits include user age groups or the historical user portraits include user age groups and user gender; Determine the number of historical movie viewing sequences corresponding to at least one historical user portrait, and determine the family structure corresponding to the device identifier to be detected based on the number of each sequence.

6. The method according to claim 5, characterized in that The determining, based on the number of each sequence, the family structure corresponding to the identification of the device to be detected includes: Based on preset selection parameters and the number of each sequence, at least one target historical user portrait is determined, and based on each target historical user portrait, the family structure corresponding to the device identifier to be detected is determined; wherein the preset selection parameters include a preset selection number or a preset number threshold; and / or, For each historical user profile, if the historical user profile corresponding to the current sequence number belongs to the middle age group, then when the current sequence number is greater than or equal to the next sequence number and greater than the previous sequence number, the historical user profile corresponding to the current sequence number is used as the target historical user profile. If the historical user profiles corresponding to the current sequence number belong to the age groups at both ends, then when the current sequence number is greater than or equal to the next sequence number or greater than the previous sequence number, the historical user profiles corresponding to the current sequence number are used as the target historical user profiles; Based on the target historical user portraits, the family structure corresponding to the device identification to be detected is determined.

7. The method according to claim 5, characterized in that The method further comprises: In response to detecting a power-on instruction corresponding to the target device identifier to be detected, obtaining recommended data based on a target family structure corresponding to the target device identifier to be detected; The recommended data is sent to a terminal device corresponding to the target device identifier to be detected, so that the terminal device displays the recommended data.

8. A user portrait generation device, characterized in that: include: a current feature vector acquisition module, configured to, in response to detecting a plurality of current movie viewing sequences, acquire a to-be-detected device identifier and a current feature vector corresponding to each of the current movie viewing sequences; A current cluster determination module is used to perform a clustering operation on multiple current feature vectors to obtain at least one current cluster; a current user portrait determination module, configured to determine the current user portrait corresponding to each current movie viewing sequence based on the current mapping list and at least one device identifier to be detected corresponding to each current cluster; The current mapping list includes at least two user portraits and multiple previous device identifiers corresponding to each of the user portraits, and the current mapping list is determined based on the previous feature vectors corresponding to the multiple previous movie viewing sequences detected last time; The current user portrait determination module is specifically used to: For each current cluster, based on at least one to-be-detected device identifier corresponding to the current cluster, determine the number of overlaps between the device identifiers corresponding to at least two user profiles in the current mapping list; The user portrait with the largest number of overlapping device identifications is used as the current user portrait of each current movie viewing sequence corresponding to the current cluster.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for generating a user portrait according to any one of claims 1 to 7.

Images