CN115545749A - E-commerce user interest analysis method and system based on artificial intelligence - Google Patents

Abstract

According to the method and the system for analyzing the user interest of the electric power company based on the artificial intelligence, user intention mining data to be analyzed can be obtained, then an interest attention knowledge set of the user intention mining data is determined through the generated user tendency variables and first user preference data of the user intention mining data, then interesting events in the user intention mining data are mined based on the interest attention knowledge set, and an interesting event mining result of the user intention mining data is determined. The user tendency variable is obtained according to the user tendency description of the interesting events in the specified data matrix, and the user tendency description among the interesting events can be represented, so that the accuracy of the user tendency description among the interesting events on the interesting event mining result can be improved in the interesting event mining process, the interesting event mining result can be accurately obtained, the user interest can be accurately analyzed, and the user experience is improved.

Description

E-commerce user interest analysis method and system based on artificial intelligence

Technical Field

The application relates to the technical field of data analysis, in particular to a commercial user interest analysis method and system based on artificial intelligence.

Background

Electronic commerce generally refers to a novel business operation mode in which, in wide commercial and trade activities worldwide, in an internet environment open on the internet, buyers and sellers conduct various commercial and trade activities without conspiracy based on a client/server application mode, and consumer online shopping, online transactions and online electronic payments among merchants, and various commercial activities, transaction activities, financial activities, and related comprehensive service activities are realized.

At present, big data is continuously developed, and the development of electronic commerce can be effectively improved when the big data is combined with electronic commerce. However, in the actual process, a user searching for a commodity determines a plurality of similar or identical search results, and the user does not spend a lot of time and effort on viewing each search result, which may result in the user not buying the commodity after going on the internet.

Disclosure of Invention

In order to solve the technical problems in the related art, the application provides a method and a system for analyzing the user interest of the electric power company based on artificial intelligence.

In a first aspect, a method for analyzing interest of a power provider user based on artificial intelligence is provided, which includes: acquiring user intention mining data to be analyzed; generating an interest attention knowledge set of the user intention mining data through the generated user tendency variable and first user preference data of the user intention mining data; wherein the user tendency variable is obtained according to the user tendency description of the interesting event in the specified data matrix; and mining the interesting events in the user intention mining data by combining the interest attention knowledge set, and generating an interesting event mining result of the user intention mining data.

In a separately implemented embodiment, the generating an interest-focused knowledge set of the user-intention-mined data from the generated user-tendency variables and the first user-preference data of the user-intention-mined data includes: compressing first user preference data of the user intention mining data to generate a preference feature compression result of the first user preference data; generating second user preference data of the user intention mining data through preference feature compression results of the first user preference data; generating an interest-interest knowledge set of the user intent mining data by the generated user trend variables, the first user preference data, and the second user preference data. Here, since the second user preference data can more clearly illustrate the user tendency description, the obtained user intention mining data can also more clearly illustrate the user tendency description, so that the determination of the interesting event mining result by the interesting attention knowledge set is more accurate, and the possibility that the interesting event mining result is interfered is reduced.

In an independently implemented embodiment, the compressing the first user preference data of the user intention mining data to generate a preference feature compression result of the first user preference data includes: and performing not less than one type of first compression processing on a plurality of first interest preference distributions of the first user preference data one by one to generate a preference characteristic compression result of the first user preference data. By performing one or more rounds of first compression processing on a plurality of first interest preference distributions of the first user preference data one by one, the user tendency description included in the first user preference data can be analyzed, and the obtained compression result of the preference characteristics of the first user preference data can enable the user tendency description to be more accurate.

In an independently implemented embodiment, the performing at least one first compression process on a plurality of first interest preference distributions of the first user preference data one by one to generate a preference feature compression result of the first user preference data includes: for one round of first compression processing in the at least one kind of first compression processing, performing compression processing on the characteristic preference compression result of the first compression unit one by one through X first compression units, and determining the characteristic compression description of the X first compression units; in an embodiment 1, in an independent implementation, the feature preference compression result of the first compression unit further includes a first interest preference distribution of the first user preference data or a feature compression description of a previous round of the first compression process. In this way, a round of the first compression process may transmit the first interest preference distribution of the first user preference data or the feature compression description of the previous round of the first compression process to the last one of the first compression units through the first compression unit, so that the feature compression description of the round of the first compression process may be more accurate.

In an independently implemented embodiment, the generating an interest-focused knowledge set of the user intent mining data from the generated user propensity variables, the first user preference data, and the second user preference data comprises: determining a user preference confidence level through the user tendency variable and the second user preference data; and splicing the first user preference data according to the user preference confidence coefficient to generate an interest attention knowledge set of the user intention mining data. The first user preference data can be analyzed in one step through the user preference confidence, so that the interest attention knowledge set obtained by splicing the first user preference data through the user preference confidence can describe the interest points of the users in the first user preference data more accurately.

In a separately implemented embodiment, the method further comprises: obtaining a specified data matrix comprising not less than one first specified data; and performing at least one second compression treatment on the at least one first designated data one by one to generate the user tendency variable. The at least one first designated data is compressed one by one in the process of carrying out second compression processing on the at least one first designated data through the artificial intelligence thread, so that the generated user tendency variable and the at least one first designated data are generated, and the user tendency variable can represent user tendency description among interested events.

In an independently implemented embodiment, said performing, one by one, not less than one second compression process on the not less than one first specified data to generate the user tendency variable includes: for one round of second compression processing in the at least one second compression processing, compressing the characteristic preference compression result of the second compression unit one by one through Y second compression units, and determining the characteristic compression description of the Y second compression unit; in an embodiment 1, in an independent implementation, the feature preference compression result of the second compression unit further includes the first specific data or a feature compression description of a previous round of the second compression processing. In this way, a round of second compression processing can transmit the first specific data or the characteristic compression description of the previous round of second compression processing to the final first compression unit through the second compression unit, so that the characteristic compression description of a round of first compression processing can be more accurate.

In an independently implemented embodiment, the mining the events of interest in the user intention mining data in conjunction with the interest attention knowledge set to generate the event of interest mining results of the user intention mining data includes: screening the significance description of the user intention mining data; and combining the significance description of the user intention mining data with the interest attention knowledge set to generate an interest event mining result of the user intention mining data. Therefore, in the process of determining the interesting event mining result of the user intention mining data, the significance description and the interesting attention knowledge set can be combined, and the credibility of the interesting event mining result is improved.

In a separately implemented embodiment, the filtering the significance description of the user intent mining data comprises: generating significance descriptions of the user intention mining data in not less than one level one by one based on the obtained second designated data; generating an event of interest mining result of the user intention mining data by combining the significance description of the user intention mining data and the interest attention knowledge set, comprising: and generating an interesting event mining result of the user intention mining data at not less than one level by combining the significance description of the user intention mining data at not less than one level and the interest attention knowledge set. Here, on the basis that there are several events of interest in the user intention mining data, the event of interest mining results may be obtained one by one according to the location (interest attention knowledge set) and the features (saliency description) of the event of interest, so that the reliability of the event of interest mining results may be improved.

In an independently implemented embodiment, the generating the significance descriptions of the user intention mining data at not less than one level one by one based on the obtained second specified data includes: performing not less than one third compression processing on the second specified data to generate a significance description of a first level in the not less than one level; and performing not less than one third compression treatment on the mining result of the interested event of the user intention mining data at the c-1 st level to generate the significance description of the user intention mining data at the c-1 st level, wherein c is an integer not less than 2. In this embodiment, the feature preference compression result of the third compression unit distributed at the front may be transmitted to the third compression unit distributed at the back, so that the feature preference compression result of the third compression unit may be cached in the database, and the obtained significance description is more accurate.

In a second aspect, an artificial intelligence based e-commerce user interest analysis system is provided, comprising a processor and a memory in communication with each other, the processor being configured to read a computer program from the memory and execute the computer program to implement the method described above.

According to the method and the system for analyzing the user interest of the electric power company based on the artificial intelligence, user intention mining data to be analyzed can be obtained, then an interest attention knowledge set of the user intention mining data is determined through the generated user tendency variable and first user preference data of the user intention mining data, then interesting events in the user intention mining data are mined based on the interest attention knowledge set, and an interesting event mining result of the user intention mining data is determined. The user tendency variable is obtained according to the user tendency description of the interesting events in the specified data matrix, and the user tendency description among the interesting events can be represented, so that the accuracy of the user tendency description among the interesting events on the mining result of the interesting events can be improved in the mining process of the interesting events, the mining result of the interesting events can be accurately obtained, the user interest can be accurately analyzed, and the experience of a user is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

Fig. 1 is a flowchart of an artificial intelligence-based consumer interest analysis method for a power company according to an embodiment of the present disclosure.

Fig. 2 is a block diagram of an artificial intelligence-based e-commerce user interest analysis apparatus according to an embodiment of the present application.

Fig. 3 is an architecture diagram of an artificial intelligence-based e-commerce user interest analysis system according to an embodiment of the present application.

Detailed Description

In order to better understand the technical solutions of the present application, the following detailed descriptions are provided with accompanying drawings and specific embodiments, and it should be understood that the specific features in the embodiments and examples of the present application are detailed descriptions of the technical solutions of the present application, and are not limitations of the technical solutions of the present application, and in a case of no conflict, the technical features in the embodiments and examples of the present application may be combined with each other.

Referring to fig. 1, a method for analyzing user interest of a commercial telephone based on artificial intelligence is shown, which may include the following steps step11-step 13.

step11, obtaining the user intention mining data to be analyzed.

step12, generating an interest attention knowledge set of the user intention mining data through the generated user tendency variable and first user preference data of the user intention mining data; wherein the user tendency variables are used to characterize a user tendency description of the event of interest.

In this embodiment, the user tendency variables for characterizing the user tendency description of the event of interest may be determined according to the user tendency description of the event of interest in the specified data matrix, for example, the specified data matrix of a set range may be obtained, and then the user tendency description of the event of interest in the specified data matrix may be filtered. The user tendency variable is associated with the location of the event of interest, and for example, the location of an event of interest to be mined in the event of interest matrix is the third event of interest location, so the user tendency variable may represent the relative location of the event of interest to be analyzed in the event of interest matrix, that is, represent the third event of interest location. To attenuate the degree of correlation of the user-preference variables with the event-of-interest characteristics, the events-of-interest in the specified data matrix may be consistent. In one possible implementation, each event of interest in the specified data matrix may also be configured to be free of characteristic information, thereby further attenuating the degree of correlation of the user-propensity variable with the characteristics of the event of interest. The feature similarity between the user tendency variables and the interesting events is low, so that the user tendency variables can be consistent or inconsistent with the user intention mining data with differences.

The first user preference data of the user intention mining data can be obtained by performing feature screening according to the user intention mining data, for example, the first user preference data of the user intention mining data can be determined by performing at least one feature extraction step on the user intention mining data through an artificial intelligence thread. And determining an interest knowledge set of the user intention mining data according to the obtained user tendency variable and the first user preference data of the user intention mining data, such as splicing the obtained user tendency variable and the first user preference data of the user intention mining data to determine the interest knowledge set of the user intention mining data. Here, since the interest attention knowledge set is obtained based on the user tendency variables and the first user preference data, the interest attention knowledge set is less disturbed by the event of interest.

step13, mining the interesting events in the user intention mining data by combining the interest attention knowledge set, and generating an interesting event mining result of the user intention mining data.

In this embodiment, the interest knowledge set may be processed through an artificial intelligence thread, for example, the interest knowledge set is subjected to recognition processing, and the like, so that an interest event mining result of the user intending to mine data may be determined. The interesting event mining result may be a mining result of mining an interesting event in the data according to a user's intention. The event of interest mining result may be an event of interest based on the user's intent mining data including an event of interest. On the basis that the interesting event matrix is included in the user intention mining data, the interesting event mining result can be an interesting event matrix, and the distribution of each interesting event in the interesting event mining result is consistent with the distribution of the corresponding interesting event in the user intention mining data.

In step12, the interest attention knowledge set of the user intention mining data can be determined through the generated user tendency variable and the first user preference data of the user intention mining data, so that the accuracy of the interest attention knowledge set is improved.

In an independently implemented embodiment, a compression process may be performed on first user preference data of the user intent mining data, a preference feature compression result of the first user preference data is determined, then second user preference data of the user intent mining data is determined based on the preference feature compression result of the first user preference data, and then an interest focus knowledge set of the user intent mining data is determined based on the specified user tendency variable, the first user preference data, and the second user preference data.

In this embodiment, the first user preference data of the user intention mining data may be compressed through the artificial intelligence thread, for example, the compression process may be performed step by step on the first user preference data, so that the user tendency description included in the first user preference data may be analyzed. However, based on the preference feature compression result obtained by compressing the first user preference data, the second user preference data of the user intention mining data can be determined, for example, the first user preference data and the preference feature compression result can be spliced to determine the second user preference data of the user intention mining data, and the second user preference data can more clearly illustrate the user tendency description compared with the first user preference data. And then, an interest knowledge set of the user intention mining data can be determined through the generated user tendency variable, the first user preference data and the second user preference data, for example, the obtained user tendency variable, the first user preference data and the second user preference data are spliced to determine the interest knowledge set of the user intention mining data, and the second user preference data can more clearly illustrate the user tendency description, so that the obtained user intention mining data can also more clearly illustrate the user tendency description, the interest event mining result determined by the interest knowledge set is more accurate, and the possibility that the interest event mining result is interfered is reduced.

For some possible embodiments, the first user preference data of the user intention mining data may be compressed, so that the user tendency description determination analysis included in the first user preference data may determine that the compression result of the preference feature of the first user preference data is more accurate.

For some alternative embodiments, the at least one first compression process may be performed on a number of first interest preference distributions of the first user preference data on a per-second basis, and the preference feature compression result of the first user preference data may be determined.

In this embodiment, the first user preference data may comprise several first interest preference distributions. The first user preference data may comprise features at several levels and the feature variables at different levels may be inconsistent. The first interest preference distribution may be a feature of the first user preference data on one level, the first compression process may be compression according to the first user preference data, further, the artificial intelligence thread may include not less than one first compression unit, and the compression process corresponding to the first compression unit may be the first compression process. Here, a round of or multiple rounds of first compression processing may be performed on a plurality of first interest preference distributions one by one through an artificial intelligence thread, a feature preference compression result of the plurality of first interest preference distributions may be determined, one first interest preference distribution may correspond to one feature preference compression result, then a plurality of feature preference compression results of a plurality of first level features may be integrated, and a preference feature compression result of the first user preference data may be generated. By performing one or more rounds of first compression processing on a plurality of first interest preference distributions of the first user preference data one by one, user tendency descriptions included in the first user preference data can be analyzed, and the obtained compression result of the preference characteristics of the first user preference data can enable the user tendency descriptions to be more accurate.

In this embodiment, for a round of first compression processing of not less than one kind of first compression processing, the feature preference compression result of the first compression unit may be compressed one by X first compression units, and the feature compression description of the X first compression units is determined; in this embodiment 1, in the embodiment, at least one type of first compression processing may be performed on the first user preference data through the artificial intelligence thread to determine a preference feature compression result of the first user preference data. The artificial intelligence thread can comprise not less than one type of first compression unit, the first compression unit can execute first compression processing, and each round of first compression processing is executed by a plurality of compression units. On the basis that the first compression process is a plurality of rounds, the steps performed by the first compression process for each round may be identical. For at least one round of first compression processing, the feature preference compression results of the round of first compression processing can be compressed one by X first compression units, one first compression unit can correspond to one feature preference compression result, and the feature preference compression results of different first compression units can be different. Further, a first compression unit may determine a feature compression description. The feature preference compression result of the first compression unit in the first round of the first compression process may be a first interest preference distribution of the first user preference data. The feature compression description of the first compression unit in the first round of the first compression process may be determined as a feature preference compression result of the first compression unit processed in the same manner in the second round of the first compression process until the last round of the first compression process. The feature compression description of the first compression unit in the last round of the first compression process may be a feature preference compression result of the first interest preference distribution described above. X first compression units may be included in a round of first compression processing, and second user preference data for which the user intends to mine data is determined. In the present embodiment, the first user preference data a, which the user intends to mine data, may be subjected to the compression process by the artificial intelligence thread. The artificial intelligence thread may include two first compression units, and each first compression unit may include a number of first compression units. Here, the first user preference data a of the user intention mining data may be loaded to the first compression unit of the artificial intelligence thread, the plurality of first interest preference distributions of the first user preference data a are compressed by the plurality of first compression units of the first compression unit, and the feature compression description of each first compression unit is determined. Wherein the feature preference compression result of the first compression unit is a first interest preference distribution, and the feature preference compression result of the second first compression unit is a feature compression description of the first compression unit and a second first interest preference distribution. Finally, the user preference data A1 of the first user preference data may be obtained. The first user preference data a and the user preference data A1 of the first user preference data may then be feature-stitched, which may be feature-consistent or feature-integrated, to determine a second user preference data of the user's intent to mine the data

In some possible implementations, the interest-focused knowledge set that the user intends to mine the data may be determined by the generated user propensity variable, the first user preference data, and the second user preference data.

For some alternative embodiments, a user preference confidence may be determined according to the obtained user tendency variable and the second user preference data, and then the interest attention knowledge set of the user intention mining data may be determined by performing a stitching process on the first user preference data according to the user preference confidence.

For some alternative embodiments, since the user preference variable and the second user preference data each include a significant user preference description, the user preference confidence may be determined based on the user preference variable and the second user preference data, such as determining a degree of association of the user preference variable with the second user preference data, and determining the user preference confidence based on the degree of association. The degree of association of the user propensity variable with the second user preference data may be determined by a weighting process of the user propensity variable with the second user preference data. And splicing the first user preference data through the obtained user preference confidence. The first user preference data can be analyzed through the user preference confidence, so that the interest attention knowledge set obtained after splicing the first user preference data through the user preference confidence can more accurately describe the interest points of the users in the first user preference data.

The user preference confidence may be determined for some possible implementations based on the obtained user propensity variable and the second user preference data. The user-propensity variables may characterize user-propensity descriptions of the events of interest, i.e., may characterize relative positioning between the events of interest.

In an embodiment of an independent implementation, a specified data matrix including not less than one first specified data may be obtained, and then not less than one second compression process may be performed on not less than one first specified data one by one, and the user tendency variable may be determined.

In this embodiment, the user tendency variable may be determined by performing one or more rounds of second compression processing on not less than one first specified data one by the artificial intelligence thread. Because not less than one first specified data is consistent, the feature matching degree between not less than one first specified data is small, and the user tendency variable and the feature matching degree obtained by performing one or more rounds of second compression processing on not less than one first specified data one by one are low. Meanwhile, in the process of carrying out second compression processing on at least one piece of first designated data through the artificial intelligence thread, at least one piece of first designated data is compressed one by one, so that the generated user tendency variable is associated with at least one piece of first designated data, namely, the user tendency variable is associated with the positioning of at least one piece of first designated data, and the user tendency variable can represent the user tendency description among interested events.

In this embodiment, for one round of second compression processing of not less than one kind of second compression processing, the feature compression description of the Y-th second compression unit may be determined by performing compression processing on the feature preference compression result of the second compression unit by Y second compression units one by one. In the embodiment 1, one or more rounds of second compression processing can be performed on not less than one first specified data one by the artificial intelligence thread to determine the user tendency variable. On the basis that the second compression process is a plurality of rounds, the steps performed by each round of the second compression process may be identical. For one round of second compression processing in at least one second compression processing, the feature preference compression results of the round of second compression processing can be compressed one by Y second compression units, one second compression unit can correspond to one feature preference compression result, and the feature preference compression results of different second compression units can be inconsistent. Further, a second compression unit may determine a feature compression description. The feature preference compression result of one second compression unit in the first round of the second compression processing may be one first specified data.

In step103, the interesting events in the user intention mining data may be mined based on the interest attention knowledge set, and the interesting event mining result of the user intention mining data is determined. In order to improve the credibility of the mining result of the event of interest, the significant description of the event of interest in the user intention mining data needs to be analyzed in the process of mining the event of interest in the user intention mining data.

In a separately implemented embodiment, the significance descriptions of the user intent mining data may be filtered and then the event of interest mining results of the user intent mining data may be determined based on the significance descriptions of the user intent mining data and the interest attention knowledge set.

In this embodiment, the significance description of the user intention mining data may be filtered, for example, the significance description of the user intention mining data may be filtered through some feature filtering threads, and then the significance description of the user intention mining data and the interest attention knowledge set are spliced to determine a splicing result. Here, the weighting coefficients of the stitching process may be configured in advance, or may be obtained by parsing according to the significance description and the interest attention knowledge set. However, based on the concatenation result, an event of interest mining result of the user's intention to mine the data may be determined, and an event of interest mining result of the user's intention to mine the data may be determined. In this way, in the process of determining the interesting event mining result of the user intention mining data, the significance description can be compared with the interesting attention knowledge set, and the credibility of the interesting event mining result is improved.

In this embodiment, the significance description of the user intention mining data in not less than one level may be determined one by one based on the obtained second specified data, and then the event of interest mining result of the user intention mining data in not less than one level may be determined based on the significance description of the user intention mining data in not less than one level and the interest attention knowledge set.

In this embodiment, the second specified data may be compressed by the artificial intelligence thread, and the significance descriptions of not less than one level may be sequentially determined, whereas the event of interest mining result of the user intention mining data of not less than one level may be determined based on the significance descriptions of the user intention mining data of not less than one level and the interest attention knowledge set of not less than one level. The significance description of one layer and the interest attention knowledge set of the same layer can correspond to the interest event mining result of one layer, and on the basis that a plurality of interest events in the user intention mining data are obtained, the interest event mining result can be obtained one by one according to the positioning (interest attention knowledge set) and the characteristics (significance description) of the interest events, so that the reliability of the interest event mining result can be improved.

In this embodiment, not less than one third compression process may be performed on the second designated data to determine the significance description of the first level in not less than one level, and then not less than one third compression process may be performed on the mining result of the event of interest of the user intention mining data at the c-1 st level to determine the significance description of the user intention mining data at the c-th level. Wherein c is an integer of not less than 2.

In this embodiment, the second specification data may be used as a feature preference compression result of not less than one third compression process in the artificial intelligence thread. Each round of third compression processing may include a plurality of third compression units, and each third compression unit may correspond to a feature preference compression result of one layer. The feature preference compression results of different third compression units may not be consistent. Further, a third compression unit may determine a feature compression description. The feature preference compression result of the first third compression unit in the first round of the third compression processing may be the second specification data. The feature compression description of the third compression unit in the first round of third compression processing may be determined as a feature preference compression result of the third compression unit processed in the same manner in the second round of third compression processing until the last round of third compression processing, and thus, not less than one third compression processing may be performed on the second specified data, and the feature compression description of the first third compression unit in the last round of third compression processing may be determined, and the feature compression description may be a saliency description of the first level in not less than one level. Further, the event mining result of interest of the first level can be determined according to the significance description of the first level and the interest attention knowledge set of the same level. The feature preference compression result of the second third compression unit in the first round of the third processing may be an event mining result of interest of the first level. And then, not less than one third compression treatment can be carried out on the mining result of the event of interest of the first layer, and the significance description of the second layer is determined. Further, event mining results of interest at the second level can be determined from the significance description at the second level and the interest attention knowledge set at the same level. Until the last round of the third compression treatment. In the last round of the third compression process, the feature compression description of the final third compression unit may be a saliency description of the final one level. Namely, the not less than one third compression processing is carried out on the mining result of the interesting event of the user intention mining data at the c-1 level, and the significance description of the user intention mining data at the c level can be determined. On the basis that c is an integer not less than 2, that is, on the basis that the third compression unit is other than the first third compression unit in the real-time third compression process, the feature preference compression result of the third compression unit may further include the feature compression description of the previous third compression unit in the round of third compression process, so that the feature preference compression result of the third compression unit distributed before may be transmitted to the third compression unit distributed after, and thus the feature preference compression result of the third compression unit may be cached in the database, so that the obtained significance description is more accurate.

For some possible embodiments, mining the events of interest in the user intention mining data in combination with the interest attention knowledge set, and generating an event of interest mining result of the user intention mining data, may further include the following steps: decomposing a user decision basis corresponding to real-time user intention data into a plurality of intention description labels according to a specified data analysis mode, and generating tendency depolarization vectors of the preference positioning of each user through an area where the preference positioning of the user is located corresponding to the optimal node of each intention description label; after generating tendency depolarization vectors of each user preference location, performing distribution processing on the tendency depolarization vectors of each user preference location to determine an average vector of each tendency depolarization vector, and determining a first interest attention knowledge set of the user decision according to the corresponding user preference location according to the average vector of each tendency depolarization vector and the tendency depolarization vector of each user preference location; in each tendency dimension, generating an interesting event corresponding to each user preference position according to whether each user preference position belongs to marginalized data, and generating a first classification result corresponding to each tendency dimension by combining a variable of each user preference position in the first interest attention knowledge set and the interesting event corresponding to each user preference position; determining a target tendency dimension corresponding to the user decision basis according to the first classification result corresponding to each tendency dimension; and generating an interesting event mining result of each user preference location through the difference in the target tendency dimension between each user preference location and the first user preference location which is most matched with the preference of the user, the difference in the target tendency dimension between the first user preference location and the second user preference location with the minimum preference of the user, and the tendency depolarization vector of each user preference location.

It can be understood that through the embodiment provided by the steps, the interesting event mining result can be determined more accurately through multi-directional analysis.

On the basis of the above, please refer to fig. 2 in combination, there is provided an artificial intelligence-based electric business user interest analyzing apparatus 200, applied to an artificial intelligence-based electric business user interest analyzing system, the apparatus comprising:

a data obtaining module 210, configured to obtain user intention mining data to be analyzed;

a knowledge set generating module 220, configured to generate an interest attention knowledge set of the user intention mining data through the generated user tendency variable and first user preference data of the user intention mining data; wherein the user tendency variable is obtained according to the user tendency description of the interesting event in the specified data matrix;

a result determining module 230, configured to mine an event of interest in the user intention mining data in combination with the interest attention knowledge set, and generate an event of interest mining result of the user intention mining data.

On the basis of the above, please refer to fig. 3 in combination, which shows an artificial intelligence-based e-commerce user interest analysis system 300, which includes a processor 310 and a memory 320, which are in communication with each other, wherein the processor 310 is configured to read a computer program from the memory 320 and execute the computer program to implement the above method.

On the basis of the above, there is also provided a computer-readable storage medium on which a computer program is stored, which when executed implements the above-described method.

In summary, based on the above scheme, user intention mining data to be analyzed may be obtained, then an interest attention knowledge set of the user intention mining data is determined through the generated user tendency variable and the first user preference data of the user intention mining data, then an interest event mining result of the user intention mining data is determined by mining an interest event in the user intention mining data based on the interest attention knowledge set. The user tendency variable is obtained according to the user tendency description of the interesting events in the specified data matrix, and the user tendency description among the interesting events can be represented, so that the accuracy of the user tendency description among the interesting events on the mining result of the interesting events can be improved in the mining process of the interesting events, the mining result of the interesting events can be accurately obtained, the user interest can be accurately analyzed, and the experience of a user is improved.

It should be appreciated that the system and its modules shown above may be implemented in a variety of ways. For example, in some embodiments, the system and its modules may be implemented in hardware, software, or a combination of software and hardware. Wherein the hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory for execution by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the methods and systems described above may be implemented using computer executable instructions and/or embodied in processor control code, for example such code provided on a carrier medium such as a diskette, CD-or DVD-ROM, programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application may be implemented not only by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but also by software executed by various types of processors, for example, or by a combination of the above hardware circuits and software (e.g., firmware).

It is to be noted that different embodiments may produce different advantages, and in different embodiments, any one or combination of the above advantages may be produced, or any other advantages may be obtained.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be considered merely illustrative and not restrictive of the broad application. Various modifications, improvements and adaptations to the present application may occur to those skilled in the art, although not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present application and thus fall within the spirit and scope of the exemplary embodiments of the present application.

Also, this application uses specific language to describe embodiments of the application. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means a feature, structure, or characteristic described in connection with at least one embodiment of the application. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, some features, structures, or characteristics of one or more embodiments of the present application may be combined as appropriate.

Moreover, those skilled in the art will appreciate that aspects of the present application may be illustrated and described in terms of several patentable species or situations, including any new and useful combination of processes, machines, manufacture, or materials, or any new and useful improvement thereon. Accordingly, various aspects of the present application may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present application may be represented as a computer product, including computer readable program code, embodied in one or more computer readable media.

The computer storage medium may comprise a propagated data signal with the computer program code embodied therewith, for example, on baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, etc., or any suitable combination. A computer storage medium may be any computer-readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code on a computer storage medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Computer program code required for the operation of various portions of the present application may be written in any one or more programming languages, including an object oriented programming language such as Java, scala, smalltalk, eiffel, JADE, emerald, C + +, C #, VB.NET, python, and the like, a conventional programming language such as C, visual Basic, fortran 2003, perl, COBOL 2002, PHP, ABAP, a dynamic programming language such as Python, ruby, and Groovy, or other programming languages, and the like. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any form of network, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service using, for example, software as a service (SaaS).

Additionally, the order in which elements and sequences of the processes described herein are processed, the use of alphanumeric characters, or the use of other designations, is not intended to limit the order of the processes and methods described herein, unless explicitly claimed. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to imply that more features are required than are expressly recited in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Where numerals describing the number of components, attributes or the like are used in some embodiments, it is to be understood that such numerals used in the description of the embodiments are modified in some instances by the modifier "about", "approximately" or "substantially". Unless otherwise indicated, "about", "approximately" or "substantially" indicates that the numbers allow for variation in flexibility. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximations that may vary depending upon the desired properties of the individual embodiments. In some embodiments, the numerical parameter should take into account the specified significant digits and employ a general digit-preserving approach. Notwithstanding that the numerical ranges and parameters setting forth the broad scope of the range are approximations, in the specific examples, such numerical values are set forth as precisely as possible within the scope of the application.

Each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, and the like, cited in this application is hereby incorporated by reference in its entirety. Except where the application history document is inconsistent or conflicting with the present application as to the extent of the present claims, which are now or later appended to this application. It is to be understood that the descriptions, definitions and/or uses of terms in the attached materials of this application shall control if they are inconsistent or inconsistent with the statements and/or uses of this application.

Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of the embodiments of the present application. Other variations are also possible within the scope of the present application. Thus, by way of example, and not limitation, alternative configurations of the embodiments of the present application may be viewed as being consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to only those embodiments explicitly described and depicted herein.

The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art to which the present application pertains. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (10)

1. An artificial intelligence-based electric business user interest analysis method is applied to a user interest analysis system, and the method at least comprises the following steps:

acquiring user intention mining data to be analyzed;

generating an interest attention knowledge set of the user intention mining data through the generated user tendency variable and first user preference data of the user intention mining data; wherein the user tendency variable is obtained according to the user tendency description of the interesting event in the specified data matrix;

and mining the interesting events in the user intention mining data by combining the interest attention knowledge set, and generating an interesting event mining result of the user intention mining data.

2. The method of claim 1, wherein generating an interest-focused knowledge set of the user intent mined data from the generated user propensity variables and the first user preference data of the user intent mined data comprises:

compressing first user preference data of the user intention mining data to generate a preference feature compression result of the first user preference data;

generating second user preference data of the user intention mining data through preference feature compression results of the first user preference data; generating an interest-interest knowledge set of the user intention mining data by the generated user tendency variable, the first user preference data and the second user preference data.

3. The method of claim 2, wherein the compressing the first user preference data for which the user intends to mine data to generate a preference feature compression result for the first user preference data comprises: and performing at least one first compression treatment on a plurality of first interest preference distributions of the first user preference data one by one to generate a preference characteristic compression result of the first user preference data.

4. The method of claim 3, wherein the performing no less than one first compression process on a number of first interest preference distributions of the first user preference data on a per-pair basis to generate preference feature compression results for the first user preference data comprises:

for one round of first compression processing in the at least one kind of first compression processing, performing compression processing on the characteristic preference compression result of the first compression unit one by one through X first compression units, and determining the characteristic compression description of the X first compression units; on the basis that a is greater than 1 and less than or equal to X, the characteristic preference compression result of the a-th first compression unit comprises the characteristic compression description of the a-1-th first compression unit, and X and a are integers greater than or equal to 1.

5. The method of claim 4, wherein the feature preference compression result of the first compression unit further comprises a first interest preference distribution of the first user preference data or a feature compression description of a previous round of the first compression process.

6. The method of any of claims 2 to 5, wherein generating the interest-focused knowledge set of user intent mining data from the generated user propensity variables, the first user preference data, and the second user preference data comprises:

determining a user preference confidence level through the user tendency variable and the second user preference data;

splicing the first user preference data according to the user preference confidence coefficient to generate an interest attention knowledge set of the user intention mining data;

wherein the method further comprises:

obtaining a specified data matrix comprising not less than one first specified data;

performing at least one second compression treatment on the at least one first designated data one by one to generate the user tendency variable;

wherein said performing at least one second compression process on said at least one first specified data one by one to generate said user propensity variable comprises:

for one round of second compression processing in the at least one second compression processing, compressing the characteristic preference compression result of the second compression unit one by one through Y second compression units, and determining the characteristic compression description of the Y second compression unit; on the basis that 1 < b is less than or equal to Y, the characteristic preference compression result of the b-th second compression unit comprises the characteristic compression description of the b-1-th second compression unit, and Y and b are integers greater than or equal to 1.

7. The method of claim 6, wherein the feature preference compression result of the second compression unit further comprises the first specified data or a feature compression description of a previous round of the second compression process.

8. The method of claim 7, wherein said mining events of interest in said user intent mining data in conjunction with said interest focus knowledge set, generating event of interest mining results for said user intent mining data, comprises: screening significance descriptions of the user intention mining data; and generating an interesting event mining result of the user intention mining data by combining the significance description of the user intention mining data and the interest attention knowledge set.

9. The method of claim 8, wherein the filtering the salient descriptions of the user's intent to mine data comprises: based on the obtained second designated data, generating significance descriptions of the user intention mining data on at least one level one by one; generating an event of interest mining result of the user intention mining data by combining the significance description of the user intention mining data and the interest attention knowledge set, comprising: generating an interest event mining result of the user intention mining data at not less than one level by combining the significance description of the user intention mining data at not less than one level and the interest attention knowledge set;

generating significance descriptions of the user intention mining data in not less than one level one by one based on the obtained second specified data, wherein the generating comprises: performing not less than one third compression processing on the second specified data to generate a significance description of a first level in the not less than one level; and performing not less than one third compression treatment on the mining result of the interesting event of the user intention mining data at the c-1 st level to generate a significance description of the user intention mining data at the c-1 st level, wherein c is an integer not less than 2.

10. An artificial intelligence based e-commerce user interest analysis system comprising a processor and a memory in communication with each other, the processor being configured to read a computer program from the memory and execute the computer program to implement the method of any one of claims 1 to 9.

Images