Method and system for determining intention target of user

Abstract

The invention belongs to the technical field of artificial intelligence, provides a method and a system for determining a user intention target, and aims to solve the problem of low accuracy of determining the user intention target in the field of artificial intelligence in the traditional technology. Whether the initial user intention target of the initial information content is clear or not is judged, if not, the expression habit style corresponding to the online user is determined, a preset information content compensation model corresponding to the expression habit style is determined, and then based on the preset information content compensation model and according to the initial information content, the initial user intention target of the initial information content is compensated. According to the method, the initial information content is subjected to information content compensation to obtain the target information content, and finally the final user intention target of the online user is determined according to the target information content, so that the accuracy of the user intention target can be improved, the utilization efficiency of online service resources is improved, and the development and prosperity of online services are facilitated.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and system for determining a user's intended target. Background Art

[0002] In the online service process, automatically determining, identifying, or predicting the user's intended goals based on artificial intelligence (AI) technologies, including but not limited to natural language processing and image recognition, is a key step in online services. User intended goals, also known as user intent, are determined starting from the moment the online user enters the online service page. This involves analyzing and determining the user's purpose or needs based on information including but not limited to historical information, browsing history, and input content (e.g., text, voice, and images). This allows the user to accurately provide the appropriate product or service to meet their purpose or needs, thereby providing online services or completing human-computer interaction. For example, on a real estate website's online property information display page, an online user browses different property listings. The user may be searching for a specific property and potentially have a need for real estate consultation. Alternatively, an online user enters "XX neighborhood," indicating that the user is seeking information about XX neighborhood and has a need to consult about it. Another example is an online user entering "XX property" on an online search service page, indicating that the user wants to learn more about XX property. Accurately understanding and providing the user's specific information about XX property is crucial. Therefore, based on the determined user intent, the more the provided content or services meet the user's actual needs or objectives, the higher the quality of the corresponding online service, and the more conducive it is to the development and prosperity of online services. Therefore, in the field of artificial intelligence, the more accurately the user intent is determined—that is, the more the user intent is determined to meet the user's actual needs—the higher the quality of the AI-based online service, that is, the faster and better the online service is completed. With the increasing frequency of online services, including but not limited to real estate information services and online search services, the demand for online services is growing, and the requirements for the accuracy of the determination of online users' user intent are becoming increasingly higher.

[0003] In online services, user intent is typically determined through various AI techniques to predict user goals or needs. For example, natural language processing can be performed on user input, identifying user intent, or predicting user intent based on historical user information, or image recognition can be used to predict user intent based on user input images.

[0004] However, the inventors recognized that conventional technologies, limited by various factors such as the complexity of online users, result in low accuracy in determining user intent using the aforementioned methods. This, in turn, makes it difficult to effectively and efficiently deliver high-quality online services, resulting in significant waste of online service resources and hindering the further development of online services. For example, in the field of real estate information services, if user intent is inaccurately determined, the property information or services recommended to users via online property information display pages may fail to meet the actual needs of online users, hindering the development and prosperity of online real estate information services.

[0005] Therefore, how to improve the accuracy of determining user intention goals has become an urgent problem that needs to be solved in the field of artificial intelligence. Summary of the Invention

[0006] The technical problem solved by the present invention is to solve the problem of low accuracy in determining user intention targets in the field of artificial intelligence.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: determine whether the online user inputs the initial information content; if the above judgment is yes, then determine whether the initial user intention target of the initial information content is clear; if the above judgment is no, determine the expression habit style of the online user; determine the preset information content compensation model corresponding to the expression habit style; based on the preset information content compensation model and according to the initial information content, compensate the initial information content to obtain the target information content; according to the target information content, determine the final user intention target of the online user.

[0008] As a preferred solution of the method for determining the user's intended target described in the present invention, determining the expression habit style of the online user includes: obtaining the historical input information corresponding to the online user; determining a number of preset historical input information groups, the preset historical input information groups including a preset expression habit style and preset other user historical input information corresponding to the preset expression habit style; based on a preset input information clustering model, clustering the personal historical input information with all the preset other user historical input information to obtain the target preset historical input information group to which the personal historical input information belongs; obtaining the target preset expression habit style corresponding to the target preset historical input information group, and obtaining the expression habit style corresponding to the online user.

[0009] The present invention also provides a system for determining a user's intended target, comprising: a first judgment module for judging whether an online user has input initial information content; a second judgment module for judging whether the initial user's intended target of the initial information content is clear if the above judgment is yes; a first determination module for determining the expression habit style of the online user if the above judgment is no; a second determination module for determining a preset information content compensation model corresponding to the expression habit style; an information content compensation module for compensating the initial information content based on the preset information content compensation model and according to the initial information content to obtain target information content; a third determination module for determining the final user's intended target of the online user based on the target information content.

[0010] The beneficial effects of the present invention are as follows: by judging whether the online user has input the initial information content, if the above judgment is yes, then judging whether the initial user intention target of the initial information content is clear, if the above judgment is no, determining the expression habit style corresponding to the online user, and determining the preset information content compensation model corresponding to the expression habit style, then based on the preset information content compensation model and according to the initial information content, the initial information content is information content compensated to obtain the target information content, and finally based on the target information content, determining the final user intention target of the online user, thereby based on the initial information content of the online user and with the help of the determined expression habit style of the online user, then based on the machine learning of the artificial intelligence corresponding to the preset information content compensation model, information content compensation is performed, and then the user intention target is determined, which can improve the semantics corresponding to the initial information content based on the most important expression habit style of the online user, and then predict and determine the user intention target, which can improve the accuracy of the user intention target, and then improve the utilization efficiency of online service resources, which is conducive to the development and prosperity of online services. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A flowchart of a method for determining a user's intended target provided by an embodiment of the present invention;

[0012] Figure 2 A schematic diagram of the overall process of the method for determining a user's intended target provided by an embodiment of the present invention;

[0013] Figure 3 A schematic diagram of a first sub-flow of the method for determining a user's intended target provided by an embodiment of the present invention;

[0014] Figure 4 A schematic diagram of a second sub-flow of the method for determining a user's intended target provided by an embodiment of the present invention;

[0015] Figure 5A schematic block diagram of a system for determining a user's intended target provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0017] An embodiment of the present invention provides a method for determining a user's intended target. The method can be applied to devices including but not limited to smartphones, tablet computers, wearable devices, terminals, computer devices, servers, cloud platforms, and the like, and can be used when determining a user's intended target in fields including but not limited to real estate information services.

[0018] In response to the technical problem of low accuracy in determining user intended targets in the field of artificial intelligence in traditional technologies, the inventors have proposed a method for determining user intended targets in accordance with an embodiment of the present invention. The core concept of this embodiment of the present invention is as follows: when an online user enters a service page, if the intended target of the user's input information is unclear, the online user's expression style is determined, and a corresponding preset information content compensation model is called to compensate the input information to obtain target information and determine the user's intended target. In the absence of input information, the online user's recent historical information is clustered to determine the concentration, frequency, and dwell time of the user's behavior trajectory, and the content with the highest concentration, the highest frequency, and the longest dwell time is selected as the user's intended target. Based on the user's input information and historical information, and with the help of artificial intelligence (including but not limited to machine learning and clustering algorithms), the online user's historical expression habits are determined, information compensation is performed, and historical information is analyzed to determine the user's intended target. This method can predict and determine the user's intended target from the perspective of the online user's most important expression style, behavior trajectory data, and psychological activities, thereby improving the accuracy of the user's intended target and thereby improving the utilization efficiency of online service resources, which is conducive to the development and prosperity of online services.

[0019] The present invention is described in detail below through specific examples.

[0020] Example 1, please refer to Figure 1 and Figure 2 , Figure 1 A flowchart of a method for determining a user's intended target provided by an embodiment of the present invention is provided. Figure 2 The overall flow chart of the method for determining the user's intended target provided by the embodiment of the present invention is as follows. Figure 1 As shown, in this embodiment, the method includes but is not limited to the following steps S11-S18:

[0021] S11. Determine whether the online user has input initial information content.

[0022] Explanatoryally, monitoring whether online users have entered initial information content on the online service page, that is, monitoring whether online users have entered information content on the online service page. The "initial" involved in the "initial information content" is only used to distinguish different information contents, and is not used to limit the information content, so that corresponding processing can be performed according to whether the online user has entered information content.

[0023] Furthermore, when an online user enters a preset service page, it is possible to monitor whether the online user has entered initial information on the online service page, thereby determining whether the online user has entered initial information in response to the online user entering the preset service page. Alternatively, when the online user stays on the preset service page, it is possible to monitor whether the online user has entered initial information on the online service page, thereby determining whether the online user has entered initial information. For example, on a real estate information service website, when an online user enters a property information display page, or when the online user stays on the property information display page, it is possible to monitor whether the online user has entered information, i.e., initial information, on the property information display page, thereby determining whether the online user has entered initial information. The input box on the property information display page is generally configured to search for related properties or real estate services. Therefore, when the online user enters initial information on the property information display page, it is generally searching for related properties or real estate services.

[0024] S12: If the online user has not input the initial information content, whether the initial user intention target of the initial information content is clear is not determined;

[0025] S13: When the online user inputs initial information content, it is determined whether the initial user intention target of the initial information content is clear.

[0026] Explanatoryally, when the online user has not input the initial information content, that is, when the online user has not input the information content on the online service page, there is no need to judge whether the initial user intention target of the initial information content is clear. When the online user inputs the initial information content, natural language processing is used to understand the initial information content and identify the initial user intention corresponding to the initial information content, that is, identify what the online user wants to do when entering the initial information content, and at the same time judge whether the initial user intention target of the initial information content is clear, that is, whether the initial user intention expressed by the initial information content is clear and complete, that is, whether the needs or purposes of the online user can be clarified through the initial information content. Since online users are different in factors including but not limited to age, education, knowledge structure, expression ability, expression habits, etc., even for the same thing, different online users generally have different ways of expression. Therefore, it is necessary to judge whether the initial user intention target corresponding to the initial information content is clear.

[0027] S14: If the initial user intended target is clear, use the initial user intended target as the final user intended target of the online user;

[0028] S15: When the initial user intention target is unclear, determine the expression habit style of the online user.

[0029] Explanatory, when the initial user intention target is clear, that is, when the initial information content clearly, completely and unambiguously expresses the needs or purposes of the online user, the initial user intention target is directly used as the final user intention target of the online user, and when the initial user intention target is unclear, that is, when the initial information content does not clearly, completely or unambiguously express the needs or purposes of the online user, the expression habit style corresponding to the online user is identified and determined based on the online user input including but not limited to the initial information content and historical input information content. Since real estate information service websites or APPs generally set the requirement for online users to log in, it is feasible to retain the historical input information content input by online users, where the expression habit style refers to the online user's The unique ways and characteristics displayed by users in the process of expressing through different modes including but not limited to text, voice, gestures, etc., the expression habit style reflects the fixed way or pattern adopted or reflected by online users when inputting information content. The formation of expression habit style is influenced by many factors such as personal life experience, educational background, and cultural literacy. The expression habit style plays an important role in interaction and can help accurately understand the expression intentions of online users. The expression habit style includes but is not limited to concise, complex, straightforward, implicit, complete, incomplete, clear, and vague, corresponding to different descriptions of the characteristics and features of online users' expression methods or patterns. Corresponding understanding and processing methods can be adopted for different expression styles to accurately understand the user's intention goals corresponding to the initial information content input by online users.

[0030] S16: Determine a preset information content compensation model corresponding to the expression habit style.

[0031] Explanatory, as described above, corresponding understanding and processing methods can be adopted for different expression styles to accurately understand the user intention target corresponding to the initial information content input by online users. Therefore, when processing and understanding the initial information content input by online users with different expression habit styles based on artificial intelligence, for each expression habit style, a corresponding information content compensation model is pre-set, that is, a preset information content compensation model. When the preset information content compensation model corresponding to each expression habit style is used to centrally process the initial information content of the corresponding category, the efficiency and quality of the initial information content processing of each expression habit style can be improved. Among them, the preset information content compensation model represents a machine learning model for compensating the initial information content, and the compensation processing of the initial information content represents that, due to The initial information content may be complex, implicit, incomplete, or ambiguous. The initial information content needs to be modified, including but not limited to simplification, directness, supplementation, and reasoning, to obtain the target information content. This allows the semantics of the target information content to be concise, direct, complete, and clear, enabling accurate and rapid understanding of the actual needs or purposes of online users, i.e., the user's intended goals. For example, complex initial information content may be simplified, incomplete initial information content may be supplemented, and ambiguous initial information content may be clarified. Preset information content compensation models include but are not limited to Transformer models, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long short-term memory networks (LSTMs). When training the preset information content compensation model corresponding to each expression habit style, training sample pairs consisting of training samples of the corresponding expression habit style and their corresponding labels are used to train the machine learning-based preset information content compensation model. This allows the preset information content compensation model to modify the training samples based on the training sample pairs and bring them closer to the corresponding labels.

[0032] Based on the above ideas and settings, after determining the expression habit style of the online user, the preset information content compensation model corresponding to the expression habit style is determined according to the expression habit style corresponding to the online user, and then the corresponding preset information content compensation model is adopted to perform corresponding compensation and modification processing on the initial information content input by the online user. This can improve the quality and efficiency of the corresponding compensation and modification processing of the initial information content input by the online user, and further improve the accuracy of understanding the initial information content, thereby improving the accuracy of understanding the user intention target corresponding to the online user.

[0033] S17: Based on the preset information content compensation model and according to the initial information content, the initial information content is compensated to obtain target information content.

[0034] Explanatoryally, according to the above description, based on the preset information content compensation model and according to the initial information content, the initial information content is subjected to corresponding compensation and modification processing to obtain the target information content. The target information content is the information expression content corresponding to the compensation and modification processing, which has the characteristics including but not limited to simplicity, straightforwardness, completeness and clarity, so as to facilitate semantic understanding, and thus accurately understand the user intention target corresponding to the online user, that is, the actual purpose or demand corresponding to the initial information content of the online user.

[0035] S18. Determine the final user intended target of the online user according to the target information content.

[0036] Explanatory, based on the target information content and then based on natural language processing, it is generally based on a preset natural language processing model, that is, a preset user intention target understanding model. The preset user intention target understanding model includes but is not limited to the Transformer model, convolutional neural network (CNN), recurrent neural network (RNN), and long short-term memory network (LSTM). Based on a unified model including but not limited to the Transformer model, convolutional neural network (CNN), recurrent neural network (RNN), and long short-term memory network (LSTM), different training samples are used for training to achieve different natural language processing functions and obtain corresponding natural language processing models. In this way, the actual needs or purposes of online users corresponding to the target information content, that is, the user intention goals, are understood, and the final user intention goals of the online users are ultimately determined. The final user intention goals are the clear, complete, and clear actual needs or purposes of the online users, so as to achieve an accurate understanding of the actual needs or purposes corresponding to the initial information content of the online users.

[0037] According to an embodiment of the present invention, whether an online user inputs initial information content is determined. If the above determination is yes, whether the initial user intention target of the initial information content is clear is determined. If the above determination is no, the expression habit style corresponding to the online user is determined, and a preset information content compensation model corresponding to the expression habit style is determined. Then, based on the preset information content compensation model and according to the initial information content, the initial information content is information content compensated to obtain the target information content. Finally, based on the target information content, the final user intention target of the online user is determined. Thus, based on the initial information content of the online user and with the help of the determined expression habit style of the online user, the information content is compensated based on the machine learning of artificial intelligence corresponding to the preset information content compensation model, and then the user intention target is determined. The semantics corresponding to the initial information content can be improved based on the most important expression habit style of the online user, and then the user intention target is predicted and determined. This can improve the accuracy of the user intention target, thereby improving the utilization efficiency of online service resources, and is conducive to the development and prosperity of online services.

[0038] In one embodiment, see Figure 3 , Figure 3 This is a schematic diagram of the first sub-flow of the method for determining the user's intended target provided by the embodiment of the present invention. Figure 3 As shown, in this embodiment, determining the expression habit style of the online user includes:

[0039] S31, obtaining the historical input information corresponding to the online user;

[0040] S32: Determine a plurality of preset historical input information groups, wherein the preset historical input information groups include a preset expression habit style and preset other user historical input information corresponding to the preset expression habit style;

[0041] S33. Clustering the user's historical input information with all preset historical input information of other users based on a preset input information clustering model to obtain a target preset historical input information group to which the user's historical input information belongs;

[0042] S34: Acquire the target preset expression habit style corresponding to the target preset historical input information group, and obtain the expression habit style corresponding to the online user.

[0043] Explanatory, a preset historical input information group is a preset historical input information group, which includes a preset expression habit style and preset other user historical input information corresponding to the preset expression habit style. Specifically, the preset other user historical input information corresponding to different preset other users can be collected, and then all the preset other user historical input information can be clustered to obtain corresponding groups. Each group is annotated with an expression habit style, and the preset expression habit style and the preset other user historical input information corresponding to the preset expression habit style can be obtained, that is, the preset historical input information group can be obtained, wherein the preset other user represents the preset other online users, and the preset other user historical input information represents the historical input information corresponding to the preset other users.

[0044] The input information clustering model is pre-set, that is, the preset input information clustering model. The preset input information clustering model represents a model for clustering input information of online users. The preset input information clustering model includes but is not limited to K-means clustering and hierarchical clustering.

[0045] Based on the above ideas and settings, please continue to refer to Figure 2 ,like Figure 2 As shown, the historical input information corresponding to the online user is obtained. The historical input information includes but is not limited to the initial information content and the historical input information content. Since the website or APP generally sets the requirement for online user login, it is possible to retain the historical input information content entered by the online user.

[0046] Then, several different preset historical input information groups are determined. The preset historical input information groups include preset expression habit styles and several preset other users' historical input information corresponding to the preset expression habit styles. Then, based on the preset input information clustering model, the user's historical input information is clustered with all preset other users' historical input information, thereby grouping the user's historical input information to obtain the target preset historical input information group to which the user's historical input information belongs. Since the expression habit style of each online user reflects the fixed method or pattern adopted or reflected when the online user inputs information content, the user's historical input information has common features and characteristics corresponding to the expression habit style. Based on the common features and characteristics, in general, the user's historical input information should also be grouped into the same preset historical input information group, thereby obtaining the target preset historical input information group to which the user's historical input information belongs. Then, the target preset expression habit style corresponding to the target preset historical input information group is obtained, which is the expression habit style corresponding to the online user, thereby obtaining the expression habit style corresponding to the online user, thereby determining the expression habit style corresponding to the online user based on the user's historical input information.

[0047] An embodiment of the present invention obtains the historical input information corresponding to an online user and determines several preset historical input information groups, wherein the preset historical input information group includes a preset expression habit style and its corresponding preset historical input information of other users, and then clusters the historical input information of the online user with all the preset historical input information of other users to obtain the target preset historical input information group to which the historical input information of the online user belongs. Based on the historical input information of different online users, the online users and their corresponding historical input information are grouped, and the target preset expression habit style corresponding to the target preset historical input information group is obtained to obtain the expression habit style corresponding to the online user. Based on the historical input information of the online user and other historical input information of other online users, and based on clustering, the expression commonalities between the online user and other online users are analyzed to classify the expression habit style of the online user, and then determine the expression habit style corresponding to the online user. Based on the historical input information of different online users and cluster analysis of all historical input information, the expression habit style of the online user is determined, which can improve the accuracy of the expression habit style corresponding to the online user, and thus improve the accuracy of the user's intended target.

[0048] In one embodiment, see Figure 4 , Figure 4 This is a second sub-flow diagram of the method for determining a user's intended target provided by an embodiment of the present invention. Figure 4 In this embodiment, the method further includes:

[0049] S41: If the online user has not input any initial information, determine a number of historical behavior information contents corresponding to the online user in a preset recent time period, wherein the historical behavior information contents include a number of historical browsing information contents corresponding to the historical behavior trajectory of the online user browsing web pages in the past and the corresponding historical browsing start time and historical browsing end time;

[0050] S42: Calculate the historical browsing duration corresponding to the online user browsing the historical browsing information content based on the historical browsing start time and the historical browsing end time;

[0051] S43, combining the historical browsing information content, the historical browsing start time, the historical browsing end time, and the historical browsing duration into a historical browsing information content set;

[0052] S44, clustering all the historical browsing information content sets based on a preset historical browsing information clustering model to obtain historical browsing information content groups;

[0053] S45, determining target historical browsing information content according to the historical browsing information content group;

[0054] S46: Determine the target history browsing information content as the user intention target corresponding to the online user.

[0055] Explanatory note: In the case that the online user has not input the initial information content, that is, in the case that the online user has not input the initial information content, it is an effective way to infer and determine the user's user intention target based on the online user's historical browsing behavior trajectory. Therefore, please continue to refer to Figure 2 ,like Figure 2 As shown, several historical behavior information contents corresponding to the online user in the preset recent time period are determined. The historical behavior information contents represent the behavior information of the historical browsing trajectory corresponding to the preset recent time period of the online user. In general, the historical behavior information contents of the preset recent time period can reflect the recent user intention target of the online user. Among them, the historical behavior information contents include several historical browsing information contents corresponding to the historical behavior trajectory of the online user's past browsing of web page content and their corresponding historical browsing start time and historical browsing end time.

[0056] Then, based on the historical browsing start time and the historical browsing end time, the historical browsing time corresponding to the online user's browsing of the historical browsing information content is calculated. The longer the historical browsing time, the more attention and attentive the online user is to the historical browsing information content, that is, the greater the interest. Then, the historical browsing information content, the historical browsing start time, the historical browsing end time, and the historical browsing time are combined into a historical browsing information content set. The historical browsing information content set can be represented by a historical browsing information content array. Since the array is arranged in an ordered sequence, using the historical browsing information content array to represent the historical browsing information content set can more effectively perform subsequent clustering.

[0057] A historical browsing information clustering model is pre-set, i.e., a preset historical browsing information clustering model. The preset historical browsing information clustering model represents a model for clustering historical browsing information content. The preset historical browsing information clustering model is not limited to K-means clustering and hierarchical clustering. Thus, based on the preset historical browsing information clustering model, all historical browsing information content sets are clustered to obtain several different historical browsing information content groups. Each historical browsing information content group is an aggregation of historical browsing information content sets with the same characteristics or features. Then, based on the obtained several historical browsing information content groups, target historical browsing information content is determined, and the target historical browsing information content is determined as the user intention target corresponding to the online user. Generally, historical browsing information content that is of great interest to online users, including but not limited to those with high attention, frequent browsing, and long stay time, is selected as the target historical browsing information content. This can accurately reflect the user intention target corresponding to the online user.

[0058] In an embodiment of the present invention, when the online user has not input initial information content, several historical behavior information contents corresponding to the online user in a preset recent time period are determined, wherein the historical behavior information contents include several historical browsing information contents corresponding to the historical behavior track of the online user browsing web content in the past and the corresponding historical browsing start time and historical browsing end time, and according to the historical browsing start time and historical browsing end time, the historical browsing time corresponding to the online user browsing the historical browsing information content is calculated, and the historical browsing information content, the historical browsing start time, the historical browsing end time, and the historical browsing time are combined into a historical browsing information content set, and then all the historical browsing information content sets are clustered to obtain a historical browsing information content set. Historical browsing information content groups are finally determined based on the historical browsing information content groups, and the target historical browsing information content is determined as the user intention target corresponding to the online user, so as to analyze the online user's historical behavior information based on the historical behavior information of the online user and with the help of artificial intelligence including but not limited to natural language processing and clustering algorithms, to achieve the determination of user intention targets based on the nearest historical behavior information, and to predict and determine user intention targets from the perspective of the online user's historical behavior trajectory data and the psychological activities it reflects, so as to improve the accuracy of user intention targets, thereby improving the utilization efficiency of online service resources, and being conducive to the development and prosperity of online services.

[0059] In one embodiment, determining target historical browsing information content according to the historical browsing information content group includes:

[0060] determining, based on a plurality of historical browsing information content sets included in the historical browsing information content group, a historical browsing attention degree of each of the historical browsing information content to the online user;

[0061] According to the historical browsing attention and the historical browsing duration, and based on a preset first sorting algorithm, all the historical browsing information contents are sorted in descending order;

[0062] Determine the historical browsing information content ranked higher as the sub-target historical browsing information content corresponding to the historical browsing information content group;

[0063] The target historical browsing information content is determined according to the sub-target historical browsing information content corresponding to all the historical browsing information content groups.

[0064] Explanatory, please continue to see Figure 2 ,like Figure 2 As shown, based on the several historical browsing information content sets contained in each historical browsing information content group, the historical browsing attention of each historical browsing information content for the online user is determined. The historical browsing attention indicates the degree of attention or interest of the online user in each historical browsing information content. The historical browsing attention is represented by, but not limited to, historical browsing frequency and historical browsing frequency. The historical browsing frequency indicates the proportion of historical browsing times of each historical browsing information content by the online user in all historical browsing times. The historical browsing frequency indicates the proportion of historical browsing occurrences. The historical browsing frequency indicates the number of times the online user has viewed each historical browsing information content in unit time. The historical browsing frequency indicates the number of times historical browsing has occurred.

[0065] A first sorting algorithm is pre-set, i.e., a preset first sorting algorithm. The preset first sorting algorithm is an algorithm for sorting all historical browsing information contents in descending order based on two factors: the historical browsing attention and the historical browsing time corresponding to the historical browsing information contents. The preset first sorting algorithm includes, but is not limited to, insertion sort, shell sort, selection sort, bubble sort, and heap sort. In addition, the "first" involved in the preset first sorting algorithm is only used to distinguish different sorting algorithms and is not used to limit the sorting algorithm. Thus, based on the historical browsing attention and the historical browsing time corresponding to the historical browsing information contents and based on the preset first sorting algorithm, all historical browsing information contents are sorted in descending order, and several historical browsing information contents with the highest ranking are determined as the sub-target historical browsing information contents corresponding to the corresponding historical browsing information content group. Then, based on the sub-target historical browsing information contents corresponding to all the historical browsing information content groups, the total target historical browsing information content of the online user is determined.

[0066] Furthermore, determining the historical browsing information content ranked higher as the sub-target historical browsing information content corresponding to the historical browsing information content group includes:

[0067] The historical browsing information content ranked first is determined as the sub-target historical browsing information content corresponding to the historical browsing information content group.

[0068] Specifically, the historical browsing information content ranked first is determined as the sub-target historical browsing information content corresponding to the historical browsing information content group, that is, after sorting the historical browsing information content group in descending order, the historical browsing information content ranked first is determined as the sub-target historical browsing information content corresponding to the historical browsing information content group. The sub-target historical browsing information content is the browsing information content with the highest attention and the greatest interest among online users in the group, which is screened out based on the two factors of historical browsing attention and historical browsing time within the corresponding historical browsing information content group. Therefore, based on the historical behavior information content of the online users, the relatively clear user intention targets of the online users in each historical browsing information content group are predicted, that is, the sub-target historical browsing information content corresponding to the historical browsing information content group, thereby improving the accuracy of determining the user intention targets based on historical behavior information analysis.

[0069] In an embodiment of the present invention, based on several historical browsing information content sets included in a historical browsing information content group, the historical browsing attention of each historical browsing information content for an online user is determined. Then, based on the historical browsing attention and historical browsing duration, all historical browsing information content is sorted in descending order based on a preset first sorting algorithm. The historical browsing information content with the highest sorting order is determined as the sub-target historical browsing information content corresponding to the historical browsing information content group. Then, based on all sub-target historical browsing information content corresponding to all historical browsing information content groups, the target historical browsing information content is determined. Thus, based on the psychological activities of the online user reflected by the historical browsing attention and historical browsing duration of each historical browsing information content, and based on the online user's interest in the historical browsing information content corresponding to the psychological activities, the sub-target historical browsing information content corresponding to each historical browsing information content group is determined. Finally, based on all sub-target historical browsing information content corresponding to all historical browsing information content groups, the target historical browsing information content is determined. This can improve the accuracy of user intention target determination based on historical behavior information analysis.

[0070] In one embodiment, determining the target historical browsing information content according to the sub-target historical browsing information content corresponding to all the historical browsing information content groups includes:

[0071] Determine the sub-goal historical browsing attention and sub-goal historical browsing duration corresponding to each of the sub-goal historical browsing information contents;

[0072] According to the sub-goal historical browsing attention and the sub-goal historical browsing time, and based on a preset second sorting algorithm, all the sub-goal historical browsing information contents are sorted in descending order;

[0073] The sub-target historical browsing information content with a higher ranking is determined as the target historical browsing information content.

[0074] Explanatoryally, in addition to the relatively clear user intention targets of online users determined in each historical browsing information content group mentioned above, that is, the sub-target historical browsing information content corresponding to the historical browsing information content group, it is also possible to screen out the total target historical browsing information content from the overall perspective of the online users based on the historical behavior information content of the online users from the overall perspective of several groups corresponding to different historical browsing information content groups. Thus, on the basis of screening out the sub-target historical browsing information content corresponding to each historical browsing information content group, the sub-target historical browsing attention and sub-target historical browsing time corresponding to each sub-target historical browsing information content are determined. The sub-target historical browsing attention is consistent with the corresponding historical browsing attention, and is a description of the same object from different perspectives, that is, both are the browsing attention of the same historical browsing information content. The sub-target historical browsing time is also consistent with the corresponding historical browsing attention, and is also a description of the same object from different perspectives, that is, both are the browsing time of the same historical browsing information content.

[0075] Therefore, please continue to see Figure 2 ,like Figure 2 As shown, according to the sub-goal historical browsing attention and the sub-goal historical browsing time, and based on the preset second sorting algorithm, from the perspective of different historical browsing information content groups, all sub-goal historical browsing information contents are sorted in descending order again, and the sub-goal historical browsing information contents with the highest sorting are determined as the target historical browsing information contents, so as to obtain the total target historical browsing information content from the overall perspective of the online user, that is, the browsing information that the online user pays the highest attention to and is of the greatest interest to the online user among all the historical browsing information contents of the online user.

[0076] In an embodiment of the present invention, by determining the sub-goal historical browsing attention and sub-goal historical browsing time corresponding to each sub-goal historical browsing information content, and then sorting all sub-goal historical browsing information contents in descending order according to the sub-goal historical browsing attention and sub-goal historical browsing time, and based on a preset second sorting algorithm, the sub-goal historical browsing information content with the highest sorting order is determined as the target historical browsing information content. Thus, from the perspective of different historical browsing information content groups, according to the psychological activities of the online user reflected by the sub-goal historical browsing attention and sub-goal historical browsing time corresponding to each sub-goal historical browsing information content, and based on the interest level of the online user in the sub-goal historical browsing information content corresponding to the psychological activities, and through secondary sorting, the target historical browsing information content is determined as the user intention target corresponding to the online user, which can further improve the accuracy of determining the user intention target based on historical behavior information analysis.

[0077] In one embodiment, after determining the sub-target historical browsing information content ranked higher as the target historical browsing information content, the method further includes:

[0078] Determine n sub-goal historical browsing information contents corresponding to the target historical browsing information contents and their corresponding descending order, wherein n is a natural number;

[0079] Based on the descending order, the n sub-goal historical browsing information contents are pushed to the online user in sequence.

[0080] Explanatory, the n sub-target historical browsing information contents corresponding to the target historical browsing information content and their corresponding descending order are determined, where n is a natural number, and based on the descending order, the n sub-target historical browsing information contents are pushed to the online user in sequence, thereby pushing them to the online user in descending order according to the online user's attention level and interest level, which can quickly meet the user's actual needs and improve the efficiency and effect of pushing information content.

[0081] The embodiment of the present invention determines n sub-target historical browsing information contents corresponding to the target historical browsing information content and their corresponding descending order, and pushes the n sub-target historical browsing information contents to the online user in sequence based on the descending order, thereby pushing the sub-target historical browsing information contents that are of high concern to the online user and that best meet the online user's expectations to the online user, which can quickly meet the actual needs of the user, improve the efficiency and effect of pushing information content, thereby meeting the actual needs of the user, and improving the attractiveness of online services to users and the customer retention rate.

[0082] In one embodiment, after determining the target history browsing information content as the user intended target corresponding to the online user, the method further includes:

[0083] Determining the target history browsing attention corresponding to the target history browsing information content;

[0084] Determining whether the target historical browsing attention is greater than or equal to a preset browsing attention threshold;

[0085] If the above judgment is yes, determine a marketing strategy corresponding to the target historical browsing information content, and push the marketing strategy to the online user;

[0086] If the above judgment is no, the marketing strategy corresponding to the target historical browsing information content is determined.

[0087] Explanatory, determine the target historical browsing attention corresponding to the target historical browsing information content. The target historical browsing attention is consistent with the above-mentioned sub-target historical browsing attention and historical browsing attention. It is a description of the same object from different angles. It is only for the convenience of describing the technical solution and is not used to limit the specific content. And judge whether the target historical browsing attention is greater than or equal to the preset browsing attention threshold. The preset browsing attention threshold represents the highest critical value of the online user's attention value to the target historical browsing information content. If the above judgment is yes, that is, the target historical browsing attention is greater than or equal to the preset browsing attention threshold, it means that the online user's attention to the target historical browsing information content has reached the highest critical value, and the online user's needs have not been met, indicating that the online user's psychological expectations have not been met. At this time, if the online user is not promptly If marketing is carried out for online users, online users may give up paying attention to the target historical browsing information content, and there is a risk of losing service customers. Therefore, the corresponding marketing strategy for the target historical browsing information content is determined. The marketing strategy represents the strategy for promoting and recommending related products or services including but not limited to real estate, and the marketing strategy is pushed to online users to provide content required by users. If the above judgment is no, that is, the target historical browsing attention is less than the preset browsing attention threshold, it means that the online user's attention to the target historical browsing information content has not reached the highest critical value. At this time, marketing can be waived for the online user and the corresponding marketing strategy for the target historical browsing information content is determined. The timing of marketing is determined based on the browsing attention level of the online user corresponding to the target historical browsing attention, so as to provide high-quality services to the online user.

[0088] In an embodiment of the present invention, the target historical browsing attention corresponding to the target historical browsing information content is determined, and whether the target historical browsing attention is greater than or equal to a preset browsing attention threshold is judged; if the above judgment is yes, it indicates that the online user has paid attention to the target historical browsing information content to a certain extent but has not yet reached a transaction, indicating that there are still factors that hinder the satisfaction of the needs of the online user. At this time, a marketing strategy corresponding to the target historical browsing information content is determined, and the marketing strategy is pushed to the online user to meet the actual needs of the user. If the above judgment is no, it indicates that the online user has not paid attention to the target historical browsing information content to a certain extent, and the intention of the online user cannot be clearly determined. At this time, the marketing strategy corresponding to the target historical browsing information content is not determined, so as to determine the timing of marketing, so as to provide high-quality services to the online user, improve the push effect of the corresponding information, and further improve the utilization efficiency of online service resources, which is conducive to the development and prosperity of online services and avoids the ineffectiveness and waste of online service resources.

[0089] It should be noted that the methods for determining user intended targets described in the above embodiments can recombine the technical features contained in different embodiments as needed to obtain a combined implementation plan, but they are all within the scope of protection required by the present invention.

[0090] In one embodiment, a system for determining a user's intended target is provided. The system for determining a user's intended target corresponds to the method for determining a user's intended target in the above embodiment. Figure 5 , Figure 5 Schematic block diagram of a system for determining a user's intended target provided by an embodiment of the present invention. Figure 5 As shown, the system 50 for determining a user's intended target includes a first judgment module 51, a second judgment module 52, a first determination module 53, a second determination module 54, an information content compensation module 55, and a third determination module 56. The above functional modules are described in detail as follows:

[0091] The first judgment module 51 is used to judge whether the online user has input the initial information content;

[0092] The second judgment module 52 is used to judge whether the initial user intention target of the initial information content is clear if the above judgment is yes;

[0093] The first determining module 53 is configured to determine the expression style of the online user if the above determination is negative;

[0094] A second determining module 54 is configured to determine a preset information content compensation model corresponding to the expression habit style;

[0095] An information content compensation module 55 is configured to compensate the initial information content based on the preset information content compensation model and the initial information content to obtain target information content;

[0096] The third determining module 56 is configured to determine the final user intended target of the online user according to the target information content.

[0097] In one embodiment, the first determining module 53 includes:

[0098] The first acquisition submodule is used to obtain the historical input information corresponding to the online user;

[0099] A first determining submodule is configured to determine a plurality of preset historical input information groups, wherein the preset historical input information groups include a preset expression habit style and preset other user historical input information corresponding to the preset expression habit style;

[0100] A first clustering submodule is configured to cluster the user's historical input information with all preset historical input information of other users based on a preset input information clustering model to obtain a target preset historical input information group to which the user's historical input information belongs;

[0101] The second acquisition submodule is used to acquire the target preset expression habit style corresponding to the target preset historical input information group, and obtain the expression habit style corresponding to the online user.

[0102] In one embodiment, the system 50 for determining the user's intended goal further includes:

[0103] a fourth determination module, configured to determine, if the online user has not input any initial information, a number of historical behavior information contents corresponding to the online user in a preset recent time period, the historical behavior information contents including a number of historical browsing information contents corresponding to the online user's historical behavior trajectory of browsing web pages in the past and the corresponding historical browsing start time and historical browsing end time;

[0104] A calculation module, configured to calculate a historical browsing duration corresponding to the online user browsing the historical browsing information content according to the historical browsing start time and the historical browsing end time;

[0105] a composition module, configured to combine the historical browsing information content, the historical browsing start time, the historical browsing end time, and the historical browsing duration into a historical browsing information content set;

[0106] A clustering module, configured to cluster all the historical browsing information content sets based on a preset historical browsing information clustering model to obtain historical browsing information content groups;

[0107] a fifth determining module, configured to determine target historical browsing information content according to the historical browsing information content group;

[0108] The sixth determining module is configured to determine the target historical browsing information content as the user intended target corresponding to the online user.

[0109] In one embodiment, the fifth determining module includes:

[0110] A second determining submodule is configured to determine the historical browsing attention of each historical browsing information content to the online user based on a plurality of historical browsing information content sets included in the historical browsing information content group;

[0111] A first sorting submodule is configured to sort all the historical browsing information contents in descending order according to the historical browsing attention and the historical browsing duration, and based on a preset first sorting algorithm;

[0112] A third determining submodule is configured to determine the historical browsing information content ranked higher as the sub-target historical browsing information content corresponding to the historical browsing information content group;

[0113] The first component submodule is configured to determine target historical browsing information content according to the sub-target historical browsing information content corresponding to all the historical browsing information content groups.

[0114] In one embodiment, the third determining submodule is specifically configured to determine the historical browsing information content ranked first as the sub-target historical browsing information content corresponding to the historical browsing information content group.

[0115] In one embodiment, the first component submodule includes:

[0116] The fourth determining submodule is used to determine the sub-goal historical browsing attention and the sub-goal historical browsing time corresponding to each of the sub-goal historical browsing information contents;

[0117] The second sorting submodule is configured to sort the contents of all the sub-goal historical browsing information in descending order according to the sub-goal historical browsing attention and the sub-goal historical browsing time, and based on a preset second sorting algorithm;

[0118] The fifth determining submodule is configured to determine the sub-target historical browsing information content with a higher ranking as the target historical browsing information content.

[0119] In one embodiment, the fifth determining module further includes:

[0120] a sixth determining submodule, configured to determine n sub-target historical browsing information contents corresponding to the target historical browsing information contents and their corresponding descending order, wherein n is a natural number;

[0121] The first push submodule is used to push the historical browsing information contents of the n sub-targets to the online user in sequence based on the descending order.

[0122] In one embodiment, the system 50 for determining the user's intended goal further includes:

[0123] a seventh determining module, configured to determine a target history browsing attention level corresponding to the target history browsing information content;

[0124] A third judgment module is used to judge whether the target historical browsing attention is greater than or equal to a preset browsing attention threshold;

[0125] The eighth determination module is used to determine the marketing strategy corresponding to the target historical browsing information content if the above judgment is yes, and push the marketing strategy to the online user

[0126] In one embodiment, the system 50 for determining the user's intended goal further includes:

[0127] The ninth determining module is configured to, when the initial user intended target is clear, use the initial user intended target as the final user intended target of the online user.

[0128] An embodiment of the present invention provides a system for determining a user's intended target, by determining whether an online user has input initial information content. If the above determination is yes, then determining whether the initial user's intended target of the initial information content is clear. If the above determination is no, determining the expression habit style corresponding to the online user, and determining a preset information content compensation model corresponding to the expression habit style. Then, based on the preset information content compensation model and according to the initial information content, the initial information content is information content compensated to obtain target information content. Finally, based on the target information content, the final user's intended target is determined for the online user. Thus, based on the online user's initial information content and with the help of the determined expression habit style of the online user, the information content is compensated based on the machine learning of artificial intelligence corresponding to the preset information content compensation model, and then the user's intended target is determined. The semantics corresponding to the initial information content can be improved based on the most important expression habit style of the online user, and then the user's intended target can be predicted and determined. This can improve the accuracy of the user's intended target, thereby improving the utilization efficiency of online service resources, which is conducive to the development and prosperity of online services.

[0129] For the specific limitations of the system for determining the user's intended target, please refer to the limitations of the method for determining the user's intended target above, which will not be repeated here. The various modules in the above-mentioned system for determining the user's intended target can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0130] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0131] The relevant data collection appearing in the embodiments of the present invention complies with the requirements of relevant laws and regulations, such as China's "Personal Information Protection Law", GDPR (EU General Data Protection Regulation) or information security standards of other countries and regions.

[0132] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for determining a user's intended target, characterized in that: include: Determine whether the online user has entered initial information; If the above judgment is yes, then determine whether the initial user intention target of the initial information content is clear; If the above judgment is negative, determining the expression style of the online user; Determining a preset information content compensation model corresponding to the expression habit style; Based on the preset information content compensation model and according to the initial information content, compensating the initial information content to obtain target information content; The final user intended target of the online user is determined according to the target information content.

2. The method for determining a user's intended target according to claim 1, wherein: Determining the online user's expression style, including: Obtaining the historical input information corresponding to the online user; Determining a plurality of preset historical input information groups, wherein the preset historical input information groups include a preset expression habit style and preset other user historical input information corresponding to the preset expression habit style; Based on a preset input information clustering model, clustering the user's historical input information with all the preset historical input information of other users to obtain a target preset historical input information group to which the user's historical input information belongs; The target preset expression habit style corresponding to the target preset historical input information group is obtained to obtain the expression habit style corresponding to the online user.

3. The method for determining a user's intended target according to claim 1 or 2, wherein: The method further comprises: If the online user does not input initial information, determine a number of historical behavior information contents corresponding to the online user in a preset recent time period, wherein the historical behavior information contents include a number of historical browsing information contents corresponding to the historical behavior track of the online user browsing web content in the past and the corresponding historical browsing start time and historical browsing end time; Calculating the historical browsing duration corresponding to the online user browsing the historical browsing information content according to the historical browsing start time and the historical browsing end time; The historical browsing information content, the historical browsing start time, the historical browsing end time, and the historical browsing duration form a historical browsing information content set; Based on a preset historical browsing information clustering model, clustering all the historical browsing information content sets to obtain historical browsing information content groups; Determining target historical browsing information content according to the historical browsing information content group; The target history browsing information content is determined as the user intention target corresponding to the online user.

4. The method for determining a user's intended target according to claim 3, wherein: Determining target historical browsing information content according to the historical browsing information content group includes: determining, based on a plurality of historical browsing information content sets included in the historical browsing information content group, a historical browsing attention degree of each of the historical browsing information content to the online user; According to the historical browsing attention and the historical browsing duration, and based on a preset first sorting algorithm, all the historical browsing information contents are sorted in descending order; Determine the historical browsing information content ranked higher as the sub-target historical browsing information content corresponding to the historical browsing information content group; The target historical browsing information content is determined according to the sub-target historical browsing information content corresponding to all the historical browsing information content groups.

5. The method for determining a user's intended target according to claim 4, wherein: Determining the historical browsing information content ranked higher as the sub-target historical browsing information content corresponding to the historical browsing information content group includes: The historical browsing information content ranked first is determined as the sub-target historical browsing information content corresponding to the historical browsing information content group.

6. The method for determining a user's intended target according to claim 5, wherein: Determining target historical browsing information content according to the sub-target historical browsing information content corresponding to all the historical browsing information content groups includes: Determine the sub-goal historical browsing attention and sub-goal historical browsing duration corresponding to each of the sub-goal historical browsing information contents; According to the sub-goal historical browsing attention and the sub-goal historical browsing time, and based on a preset second sorting algorithm, all the sub-goal historical browsing information contents are sorted in descending order; The sub-target historical browsing information content ranked higher is determined as the target historical browsing information content.

7. The method for determining a user's intended target according to claim 6, wherein: After determining the sub-target historical browsing information content ranked first as the target historical browsing information content, the method further includes: Determine n sub-goal historical browsing information contents corresponding to the target historical browsing information contents and their corresponding descending order, wherein n is a natural number; Based on the descending order, the n sub-goal historical browsing information contents are pushed to the online user in sequence.

8. The method for determining a user's intended target according to claim 3, wherein: After determining the target historical browsing information content as the user intended target corresponding to the online user, the method further includes: Determining the target history browsing attention corresponding to the target history browsing information content; Determining whether the target historical browsing attention is greater than or equal to a preset browsing attention threshold; If the above judgment is yes, a marketing strategy corresponding to the target historical browsing information content is determined, and the marketing strategy is pushed to the online user.

9. The method for determining a user's intended target according to claim 1 or 2, wherein: The method further comprises: In the case that the initial user intention target is clear, the initial user intention target is used as the final user intention target of the online user.

10. A system for determining a user's intended goal, characterized in that: include: The first judgment module is used to judge whether the online user has input the initial information content; A second judgment module is used to judge whether the initial user intention target of the initial information content is clear if the above judgment is yes; A first determining module is configured to determine the expression style of the online user if the above determination is negative; A second determining module is used to determine a preset information content compensation model corresponding to the expression habit style; an information content compensation module, configured to compensate the initial information content based on the preset information content compensation model and the initial information content to obtain target information content; The third determining module is configured to determine the final user intended target of the online user according to the target information content.

Images