CN112084376A - Map knowledge based recommendation method and system and electronic device - Google Patents

Abstract

The invention discloses a recommendation method, a recommendation system and an electronic device based on atlas knowledge, wherein the recommendation method based on atlas knowledge comprises the following steps: step S1: acquiring data provided by a user and performing data arrangement; step S2: extracting map elements from the sorted data and constructing knowledge map data based on the map elements; step S3: performing model training according to the knowledge graph data; step S4: and obtaining a recommendation result through the trained model according to the real-time data. The method and the system show the relevance of the prior behaviors and the results in a relevance relation mode, can clearly check that the relation between certain behaviors and the entity is relatively tight, and can bind the tight relation to perform relevant adjustment in a targeted manner in a subsequent target optimization scene.

Description

Map knowledge based recommendation method and system and electronic device

Technical Field

The present invention relates to a recommendation method, a recommendation system, and an electronic device, and in particular, to a recommendation method, a recommendation system, and an electronic device for displaying knowledge graph interpretations based on group users.

Background

With the rapid development of networks and the rapid expansion of information, the rhythm of user behaviors is accelerated, more and more information which is directly required or potentially required by a user can be rapidly acquired, and the application of a recommendation system under the requirement is more and more extensive.

The client needs to view various data indexes of the recommended position and also expects to view the association relation in the recommendation process. And a clear incidence relation display is constructed through the knowledge graph, the rationality of the recommended content is checked, and a deeper insight conclusion is obtained through analysis of the incidence relation.

In the prior art, data indicators of different behaviors (such as exposure, click, purchase, praise, and the like) are shown in a data report corresponding to a recommendation bit. However, in practice it has been found that the following disadvantages are present:

1. the tables of data metrics must correspond exactly in time and cannot exhibit behavior sequence features and associations. The fracture review behavior data may produce omissions in the data phenomenon or draw erroneous conclusions.

2. And (3) converting the characteristics and behaviors of the user into labels, and displaying the dimension data of a single label or a plurality of labels.

3. There is no presentation of time sequence or association between labels, and it is impossible to progress layer by layer on the data display, resulting in the effect of analyzing from a single dimension down.

Therefore, there is an urgent need to develop a recommendation method, a recommendation system and an electronic device based on map knowledge, which overcome the above-mentioned drawbacks.

Disclosure of Invention

In order to solve the above problems, the present invention provides a recommendation method based on atlas knowledge, wherein the recommendation method comprises:

step S1: acquiring data provided by a user and performing data arrangement;

step S2: extracting map elements from the sorted data and constructing knowledge map data based on the incidence relation among the map elements;

step S3: performing model training according to the knowledge graph data;

step S4: and obtaining a recommendation result through the trained model according to the real-time data.

The method for recommending based on the map knowledge further comprises the following steps:

step S5: and outputting a single user association track by combining the historical behaviors of the users and the recommendation result.

The method for recommending based on the map knowledge further comprises the following steps:

step S6: and extracting data of each user for aggregation statistics and outputting a statistical result.

The above recommendation method based on atlas knowledge, wherein the data includes: user data and material data, the user data including user behavior and user characteristics.

In the above recommendation method based on map knowledge, the step S1 includes defining a tag structure for the data.

In the above recommendation method based on map knowledge, step S2 includes:

step S21: processing the data through word segmentation and semantic processing;

step S22: performing entity extraction, relationship extraction and event extraction;

step S23: and the labels, the association relation and the entities are corresponded to form knowledge graph data.

In the above recommendation method based on atlas knowledge, in step S5, the historical behaviors of the user referred to by the model are selected, the classifications of different behaviors of the user are recorded at the same time, and a single user association track is output according to the historical behaviors, the classifications and the recommendation result.

In the above recommendation method based on map knowledge, step S6 includes extracting user behaviors and user characteristics of each user, performing aggregation statistics with different characteristics as dimensions, and outputting the statistical result.

The invention also provides a recommendation system based on the atlas knowledge, which comprises the following components:

a data processing unit: acquiring data provided by a user and performing data arrangement;

a construction unit: extracting map elements from the sorted data and constructing knowledge map data based on the incidence relation among the map elements;

a training unit: performing model training according to the knowledge graph data;

a result output unit: obtaining a recommendation result through the trained model according to the real-time data;

a trajectory output unit: outputting a single user association track by combining the historical behaviors of the users and the recommendation result;

a statistical result output unit: and extracting data of each user for aggregation statistics and outputting a statistical result.

The invention also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the graph knowledge-based recommendation method in any one of the above items through the computer program.

In summary, compared with the prior art, the invention has the following effects: the method can use the marks of the entity relationship of the knowledge graph and the results with strong association relationship to summarize the behavior characteristics of a single user, summarize the common behavior tracks and behavior characteristics of group users, and use the association relationship which is displayed clearly in a graphical way, thereby being convenient for rapidly finding data phenomena and analyzing the data; meanwhile, based on the diversity and complexity of the association relationship, the relationship graph structure can be displayed with emphasis.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

FIG. 1 is a flow chart of a recommendation method of the present invention;

FIG. 2 is a flowchart of step S2 in FIG. 1;

FIG. 3 is a schematic diagram illustrating the application of the recommendation method of the present invention;

FIG. 4 is a diagram illustrating a recommendation method according to the present invention;

FIG. 5 is a schematic diagram of a recommendation system of the present invention;

fig. 6 is a schematic diagram of a hardware structure of the electronic device according to the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The exemplary embodiments and descriptions of the present invention are provided to explain the present invention and not to limit the present invention. Additionally, the same or similar numbered elements/components used in the drawings and the embodiments are used to represent the same or similar parts.

As used herein, the terms "first", "second", "S1", "S2", …, etc. do not particularly denote an order or sequential meaning, nor are they intended to limit the present invention, but merely distinguish between elements or operations described in the same technical terms.

With respect to directional terminology used herein, for example: up, down, left, right, front or rear, etc., are simply directions with reference to the drawings. Accordingly, the directional terminology used is intended to be illustrative and is not intended to be limiting of the present teachings.

As used herein, the terms "comprising," "including," "having," "containing," and the like are open-ended terms that mean including, but not limited to.

As used herein, "and/or" includes any and all combinations of the described items.

References to "plurality" herein include "two" and "more than two"; reference to "sets" herein includes "two sets" and "more than two sets".

As used herein, the terms "substantially", "about" and the like are used to modify any slight variation in quantity or error that does not alter the nature of the variation. Generally, the range of slight variations or errors modified by such terms may be 20% in some embodiments, 10% in some embodiments, 5% in some embodiments, or other values. It should be understood by those skilled in the art that the aforementioned values can be adjusted according to actual needs, and are not limited thereto.

Certain words used to describe the present application are discussed below or elsewhere in this specification to provide additional guidance to those skilled in the art in describing the present application.

Knowledge map (Knowledge Graph) is a series of different graphs displaying Knowledge development process and structure relationship in the Knowledge domain visualization or Knowledge domain mapping map in the book intelligence world, and is used for describing Knowledge resources and carriers thereof by using visualization technology, mining, analyzing, constructing, drawing and displaying Knowledge and mutual relation among the Knowledge resources and the carriers.

The knowledge graph is a modern theory which achieves the aim of multi-discipline fusion by combining theories and methods of applying subjects such as mathematics, graphics, information visualization technology, information science and the like with methods such as metrology introduction analysis, co-occurrence analysis and the like and utilizing a visual graph to vividly display the core structure, development history, frontier field and overall knowledge framework of the subjects. It can provide practical and valuable reference for subject research.

The knowledge graph is a data structure based on a graph and consists of nodes (points) and edges (edges), wherein each node represents an entity, each Edge is a relation between the entities, and the knowledge graph is a semantic network in nature. An entity refers to something in the real world, such as a person, place name, company, phone, animal, etc.; relationships are used to express some kind of linkage between different entities. From the top, it can be seen that the entities have place names and people; the major theory belongs to Yunnan, the Xiaoming lives in the major theory, the Xiaoming and the Xiaoqin are friends, and all are relationships among entities. Popular definition: a knowledge graph is a relational network that links together all of the different kinds of information, and thus provides the ability to analyze problems from a "relational" perspective.

According to the recommendation method based on the map knowledge, after a knowledge map is constructed in the early stage and an applicable model is trained, the behavior data of the user can be automatically combined, the behavior track of the user can be output and stored, the key dimensions can be extracted, the summary statistics can be carried out aiming at the key dimensions, the key entity association relation can be displayed in different dimensions, and the range of the different dimensions can be defined for detailed viewing analysis.

Its nodes represent entities (entries) or concepts (concepts), and edges represent various semantic relationships between entities/concepts.

Referring to fig. 1, fig. 1 is a flowchart illustrating a recommendation method according to the present invention. As shown in fig. 1, the graph knowledge-based recommendation method of the present invention includes:

step S1: and acquiring data provided by a user and performing data arrangement. Wherein the step S1 includes defining a tag structure for the data. Specifically, the data includes: user data and material data, the user data including user behavior and user characteristics. And (4) performing data sorting according to user data and material data provided by a client, wherein the provided data can be directly used if a clear and regular label is provided, and the provided data needs to be reprocessed if the label is not clear. The tag structure is defined for the data first: for example, determining how many labels to distinguish, the course may have labels of instructor, price, classification, etc.; for example, what the dictionary value for each tag is and what range, such as a range of 10 years for a particular age of the user.

Step S2: and extracting map elements from the sorted data and constructing knowledge map data based on the incidence relation among the map elements.

Referring to fig. 2, fig. 2 is a flowchart of step S2 in fig. 1. As shown in fig. 2, the step S2 includes:

step S21: processing the data through word segmentation and semantic processing;

step S22: performing entity extraction, relationship extraction and event extraction;

step S23: and the labels, the association relation and the entities are corresponded to form knowledge graph data.

Specifically, in this step, the information such as the title and the introduction may be processed by word segmentation and semantic processing, for example, if the basic information exists in the link, the basic information may be obtained by a crawler, and then text processing may be performed to associate the tag, the association relationship, and the like with the entity.

And (3) extracting the relation:

defining: certain semantic relationships that exist between entities are automatically identified. The method can be divided into binary relation extraction (two entities) and multivariate relation extraction (three or more entities) according to the number of the participating entities.

By focusing on the semantic relationship between two entities, one can get (arg1, relationship, arg2) triples, where arg1 and arg2 represent the two entities and the relationship represents the semantic relationship between the entities. (e.g., semantic relationships between words in a sentence can be obtained through the Hanlp analysis tool).

The relation extraction method comprises the following steps:

the method for extracting the limited domain relation comprises the following steps:

the method for extracting the relation based on the template comprises the following steps: the template obtained through manual editing or learning is used for extracting and distinguishing entity relations in the text, the template quality and the coverage are limited, and the expandability is not strong. (the court document of doing oneself belongs to the extraction based on the template)

The relation extraction method based on machine learning comprises the following steps: relationship extraction is considered a classification problem, wherein the relationship extraction method based on machine learning can be divided into supervised and weakly supervised methods.

The supervised relationship extraction method comprises the following steps:

the method based on the characteristic engineering comprises the following steps: explicit conversion of relationship instances into classifier-acceptable feature vectors

Kernel function based approach: directly taking the structure tree as a processing object, and using a kernel function instead of an inner product of feature vectors when calculating the distance between the relations

Neural network based methods: automatic learning of efficient feature representations directly from input text, end-to-end

The weak supervision relation extraction method comprises the following steps: and a large amount of data does not need to be marked manually.

Distance supervision: the open knowledge graph is used for automatically marking the training samples without manual one-by-one marking, and belongs to one of weak supervision relation extraction.

The method for extracting the open domain relation comprises the following steps:

the relationship categories need not be predefined, and some words in the entity-pair context are used to describe the relationships between entities.

The event extraction and method comprises the following steps:

defining: events of interest to the user are extracted from the text describing the event information and presented in a structured form.

The method comprises the following steps: firstly, the event and its type are identified, secondly, the elements (generally, entities) involved in the event are identified, and finally, the role each element plays in the event needs to be determined.

Event extraction related concepts:

event designation: description in natural language of a particular event occurring objectively, usually a sentence or a group of sentences

Event trigger words: the word which can represent the event occurrence in the event reference is an important characteristic for determining the event category, and is generally a verb or noun

Event element: participants in an event, consisting essentially of entities, time and attribute values

Element roles: what role an event element plays in a corresponding event

Event types are as follows: event elements and trigger words determine the category of the event (a category defines several subcategories)

And (3) limited domain event extraction: before extraction, the types of target events and the specific structure of each type (which includes specific event elements) are predefined, and a certain amount of label data is usually given.

The extraction method of the limited domain event comprises the following steps:

the method based on pattern matching: the identification and extraction of certain type of event is carried out under the guidance of some patterns (step: pattern acquisition, pattern matching)

Supervised event pattern matching: obtaining of patterns is based entirely on corpora labeled manually

Weakly supervised event pattern matching: the corpus is not required to be completely marked, and only a certain pre-classification or a small quantity of seed modes are manually made on the corpus

Method based on machine learning

The supervised event extraction method comprises the following steps: modeling event extraction as a multi-classification problem

The method based on the characteristic engineering comprises the following steps: the event instances need to be explicitly converted into characteristic vectors which can be accepted by a classifier, and research is focused on how to extract distinctive characteristics

Neural network based methods: automatically acquire features from the text to extract events, thereby avoiding the problem of error accumulation caused by using the traditional natural language processing tool

And (3) entity extraction:

entity extraction or Named Entity Recognition (NER) plays an important role in information extraction, mainly extracting atomic information elements in text, such as name of person, organization/organization name, geographical location, event/date, character value, denomination value, etc. The entity extraction task has two keywords: find & classify, find named entity, and classify.

The method for extracting the entity is divided into 3 types, wherein the rule-based method generally needs to write a template for a target entity and then carry out matching in the original corpus; the method based on statistical machine learning is mainly characterized in that an original corpus is trained through a machine learning method, and then an entity is identified by utilizing a trained model; extraction oriented to the open domain is oriented to mass Web corpora.

Step S3: and carrying out model training according to the knowledge graph data. Specifically, the user data and material data input model is used for training the model to obtain a recommendation result of initial access.

Step S4: and obtaining a recommendation result through the trained model according to the real-time data. Specifically, in this step, the trained model is online and then output of the recommendation result in combination with the real-time user behavior data, that is, the recommendation result is output through the continuously supplemented new user and the real-time data of the user in combination with the real-time user behavior data, for example, the user can be recommended the mother and infant related commodities according to the fact that the user newly searches the mother and infant commodities within 1 hour.

Step S5: and outputting a single user association track by combining the historical behaviors of the users and the recommendation result. Specifically, in this step, since the recommendation result of the user is a black box for the client, the content can be converted into interpretable content, the historical behavior of the user referred by the model is selected, the categories of different behaviors of the user are recorded, for example, the user a purchases a financial class course (the price range, the duration range, the instructor information, and the like of the financial class course are expanded), searches an art class course (the price range, the duration range, the instructor information, and the like of the art class course are also expanded), and clicks the art class course (whether the click/other behavior is the recommended commodity for the user is identified).

The user behavior track is analyzed, the user preference is researched, and technical support is provided for an active service mode of 'information searching user'. According to the analysis result of the user behavior track, user preference information is obtained, contents which accord with the preference of the user are recommended to the user, and active service of 'information searching user' is achieved. According to the evaluation mechanism recommended by the content, the evaluation of the user on the active service is conveniently collected, and the optimization of a follow-up behavior trajectory analysis model is facilitated. And circularly optimizing the trajectory analysis model, and feeding the optimization result back to the user to realize sustainable accurate active service of a closed loop.

The application scene is as follows:

(1) and recommending the application content. In the application process, a user finds that the dimension condition is too many, the frequently used dimension is not the top, the condition value needs to be reselected every time, and the operation is inconvenient. Through analysis, the user is found to frequently use only a few dimensions of home city, time, brand, and the like. The solution measures are as follows: the dimensions frequently used by the user are recommended and displayed, the dimensions not frequently used are hidden, the recommended operation sequence can automatically perform typesetting of dimension display according to the dimension sequence frequently operated by the user, and the condition values frequently selected or input by the user are automatically recommended and filled.

(2) Focus on application recommendations. The user finds that the system is too large and the number of functional nodes is too large in the application process, the user needs to click multiple layers each time to find the functional pages needed to be used, and concerned applications are scattered in different places and are inconvenient to use. According to the application model concerned by the user, the applications frequently visited by the user in daily work are analyzed, the fact that the number of functional nodes is too large and the tree node on the left side of the system is too large is found, so that the user is inconvenient to use, the applications concerned by the user are scattered in different places and are inconvenient to use, and only a plurality of KPIs and ad hoc query applications are frequently used by the user. The solution measures are as follows: KPIs and on-demand query applications frequently used by users are actively recommended to personalized office desktops of the users, or collectible folders, or home pages and other places where the users can directly use the applications, so that the users can use the applications conveniently.

(3) And (5) auxiliary key application guarantee. The user finds some key applications in the application process, the user feedback is frequently repeatedly inquired for many times, and data cannot be found. And confirming the key application according to the user attention application model and the system optimization model, and analyzing the centralized access time period of the key application. The solution measures are as follows: and the background UTAP adjusts the priority of the task according to the access time requirement of the user, and ensures that the priority execution of the key application is guaranteed.

Step S6: and extracting data of each user for aggregation statistics and outputting a statistical result. Specifically, in this step, the user and behavior characteristics of each user are extracted, statistics are aggregated by taking different characteristics as dimensions (for example, the aggregated data is viewed by taking regions, categories, price ranges, etc. as main dimensions), according to the result of step S5, the user 'S labels, such as the region information of the individual user' S behavior, the user 'S gender, age and other demographic information, the user' S occupation and other work information, are supplemented, and then the data of the user 'S behavior and labels are summarized to obtain the association relationship between the user' S label and the recommended content, thereby obtaining the conditions of the group user recommendation reason and recommendation effect, being displayed on the interface in a visual way for the client to know the behavior preference and recommendation effect of the user, the system can guide the formulation of the operation strategy and can assist in increasing the popularization activity in a targeted manner.

Referring to fig. 3, fig. 3 is a schematic application diagram of the recommendation method of the present invention. As shown in fig. 3, the graph knowledge-based recommendation method of the present invention is applied to roughly the following processes:

1. acquiring data provided by a client for data arrangement;

2. performing map construction, including entity extraction, relation extraction, event extraction and the like;

3. carrying out model training by using the constructed data;

4. outputting a recommendation result by combining real-time user behavior data after the trained model is online;

5. outputting a single user association track by combining the historical behaviors of the users and the recommendation result;

6. the user and behavior characteristics of each user are extracted, and different characteristics are taken as dimension convergence statistics (for example, regions, categories, price ranges and the like are taken as main dimensions to check summarized data).

Referring to fig. 4, fig. 4 is a diagram illustrating a recommendation method of the present invention. As shown in FIG. 4, the recommendation method of the present invention shows the relevance of the previous behavior and the result in a relevance relationship manner, and can clearly view the relationship between some behaviors and entities.

In another embodiment of the invention, a label marking mode can be used, and a plurality of label dimensions are combined to check possible association relations.

Referring to fig. 5, fig. 5 is a schematic structural diagram of a recommendation system of the present invention. As shown in fig. 5, the recommendation system of the present invention includes:

the data processing unit 11: acquiring data provided by a user and performing data arrangement, wherein a data processing unit 11 defines a tag structure for the data;

the construction unit 12: extracting map elements from the sorted data and constructing knowledge map data based on the association relationship among the map elements;

the training unit 13: performing model training according to the knowledge graph data;

the result output unit 14: obtaining a recommendation result through the trained model according to the real-time data;

trajectory output unit 15: outputting a single user association track by combining the historical behaviors of the user and the recommendation result, wherein the track output unit 15 selects the historical behaviors of the user referenced by the model and records the classification of different behaviors of the user at the same time, and outputting the single user association track according to the historical behaviors, the classification and the recommendation result;

statistical result output unit 16: and extracting data of each user, performing convergence statistics and outputting a statistical result, wherein the statistical result output unit 16 extracts user behaviors and user characteristics of each user, performs convergence statistics by taking different characteristics as dimensions, and outputs the statistical result.

Wherein the data comprises: user data and material data, the user data including user behavior and user characteristics.

Further, after the construction unit 12 processes the data through word segmentation and semantic processing, the construction unit 12 performs entity extraction, relationship extraction and event extraction; then, the construction unit 12 corresponds the tag, the association relationship and the entity to form knowledge graph data.

The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, and wherein the program executes to perform any of the graph knowledge-based recommendation methods described above.

The invention also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the graph knowledge-based recommendation method in any one of the above items through the computer program.

Specifically, the entity recommendation method of the embodiment of the present application described in conjunction with fig. 1 may be implemented by an electronic device. Fig. 6 is a schematic diagram of a hardware structure of the electronic device according to the present invention.

The electronic device may include a processor 81 and a memory 82 storing computer program instructions.

Specifically, the processor 81 may include a Central Processing Unit (CPU), or A Specific Integrated Circuit (ASIC), or may be configured to implement one or more Integrated circuits of the embodiments of the present Application.

Memory 82 may include, among other things, mass storage for data or instructions. By way of example, and not limitation, memory 82 may include a Hard Disk Drive (Hard Disk Drive, abbreviated to HDD), a floppy Disk Drive, a Solid State Drive (SSD), flash memory, an optical Disk, a magneto-optical Disk, tape, or a Universal Serial Bus (USB) Drive or a combination of two or more of these. Memory 82 may include removable or non-removable (or fixed) media, where appropriate. The memory 82 may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory 82 is a Non-Volatile (Non-Volatile) memory. In particular embodiments, Memory 82 includes Read-Only Memory (ROM) and Random Access Memory (RAM). The ROM may be mask-programmed ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), Electrically rewritable ROM (Electrically Alterable Read-Only Memory, EEPROM), or FLASH Memory (FLASH), or a combination of two or more of these, where appropriate. The RAM may be a Static Random-Access Memory (SRAM) or a Dynamic Random-Access Memory (DRAM), where the DRAM may be a Fast Page Mode DRAM (Fast Page Mode Dynamic Random Access Memory, FPMDRAM), an Extended data output DRAM (Extended data Access Memory, EDODRAM), a Synchronous DRAM (Synchronous Dynamic Random-Access Memory, SDRAM), and so on.

The memory 82 may be used to store or cache various data files for processing and/or communication use, as well as possible computer program instructions executed by the processor 81.

The processor 81 reads and executes the computer program instructions stored in the memory 82 to implement any one of the recommendation methods in the above embodiments.

In some of these embodiments, the electronic device may also include a communication interface 83 and a bus 80. As shown in fig. 6, the processor 81, the memory 82, and the communication interface 83 are connected via the bus 80 to complete communication therebetween.

The communication interface 83 is used for realizing communication among modules, devices, units and/or equipment in the embodiment of the present application. The communication port 83 may also be implemented with other components such as: data communication is carried out among external equipment, image/data acquisition equipment, a database, external storage, an image/data processing workstation and the like.

The bus 80 includes hardware, software, or both to couple the components of the electronic device to one another. Bus 80 includes, but is not limited to, at least one of the following: data Bus (Data Bus), Address Bus (Address Bus), Control Bus (Control Bus), Expansion Bus (Expansion Bus), and Local Bus (Local Bus). By way of example, and not limitation, Bus 80 may include an Accelerated Graphics Port (AGP) or other Graphics Bus, an Enhanced Industry Standard Architecture (EISA) Bus, a Front-Side Bus (Front Side Bus), an FSB Bus, a Hyper Transport (HT) Interconnect, an ISA Bus, an InfiniBand (InfiniBand) Interconnect, a Low Pin Count (LPC) Bus, a memory Bus, a microchannel Architecture (MCA) Bus, a PCI-Interconnect (Peripheral Component Interconnect) Bus, a PCI-Express (PCI-X) Bus, a Serial Technology Attachment (Serial Technology Attachment, SATA) Bus, a Local Electronics Association (Video Association), abbreviated VLB) bus or other suitable bus or a combination of two or more of these. Bus 80 may include one or more buses, where appropriate. Although specific buses are described and shown in the embodiments of the application, any suitable buses or interconnects are contemplated by the present application.

The electronic device can clearly check that the relationship between certain behaviors and entities is relatively tight by showing the relationship between the previous behaviors and the result in a manner of the association relationship, and can bind the tight relationship to perform relevant adjustment in a targeted manner in a subsequent target optimization scene, so that the recommendation method described in combination with fig. 1 is realized.

In summary, the invention shows the relevance between the previous behavior and the result in the way of the relevance relationship, can clearly check that the relationship between some behaviors and entities is relatively tight, and can bind the tight relationship to perform relevant adjustment in a targeted manner in a subsequent target optimization scene.

Although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A recommendation method based on atlas knowledge is characterized by comprising the following steps:

step S1: acquiring data provided by a user and performing data arrangement;

step S2: extracting map elements from the sorted data and constructing knowledge map data based on the incidence relation among the map elements;

step S3: performing model training according to the knowledge graph data;

step S4: and obtaining a recommendation result through the trained model according to the real-time data.

2. The graph-knowledge-based recommendation method of claim 1, further comprising:

step S5: and outputting a single user association track by combining the historical behaviors of the users and the recommendation result.

3. The graph-knowledge based recommendation method of claim 2, further comprising:

step S6: and extracting data of each user for aggregation statistics and outputting a statistical result.

4. The graph-knowledge based recommendation method of claim 3, wherein the data comprises: user data and material data, the user data including user behavior and user characteristics.

5. The graph knowledge-based recommendation method according to any one of claims 1-4, wherein said step S1 includes defining a tag structure for said data.

6. The graph knowledge-based recommendation method according to claim 5, wherein the step S2 comprises:

step S21: processing the data through word segmentation and semantic processing;

step S22: performing entity extraction, relationship extraction and event extraction;

step S23: and the labels, the association relation and the entities are corresponded to form knowledge graph data.

7. The method for recommending based on atlas knowledge according to claim 2, wherein the step S5 includes selecting the historical behaviors of the user referenced by the model and recording the classification of different behaviors of the user at the same time, and outputting a single user association track according to the historical behaviors, the classification and the recommendation result.

8. The graph knowledge-based recommendation method according to claim 3, wherein the step S6 comprises extracting user behaviors and user characteristics of each user, performing aggregation statistics with different characteristics as dimensions, and outputting the statistical results.

9. A graph knowledge-based recommendation system, comprising:

a data processing unit: acquiring data provided by a user and performing data arrangement;

a construction unit: extracting map elements from the sorted data and constructing knowledge map data based on the incidence relation among the map elements;

a training unit: performing model training according to the knowledge graph data;

a result output unit: obtaining a recommendation result through the trained model according to the real-time data;

a trajectory output unit: outputting a single user association track by combining the historical behaviors of the users and the recommendation result;

a statistical result output unit: and extracting data of each user for aggregation statistics and outputting a statistical result.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the graph knowledge-based recommendation method according to any one of claims 1 to 8 by the computer program.

Images