Knowledge graph-based package recommendation method, apparatus and device, and storage medium

Abstract

The invention provides a package recommendation method and device based on a knowledge graph, equipment and a storage medium, and the package recommendation method based on the knowledge graph comprises the steps: obtaining input package information, determining a first package according to the package information, determining at least one second package associated with the first package from the knowledge graph according to the first package, and recommending the at least one second package to the knowledge graph according to the at least one second package. The associated second package can be automatically determined by using the information of the first package based on the knowledge graph, which is more efficient; similarity analysis is conducted on the first package and the second package, the similarity result of the first package and the second package is determined, similarity comprises label similarity analysis and quantity similarity, and the accuracy of similarity analysis is improved; and determining the target package according to the similarity result, and recommending the target package, thereby reducing the dependence of human experience, facilitating the improvement of the recommendation effect and efficiency, and facilitating the improvement of the marketing effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of computers, and in particular to a package recommendation method and device based on a knowledge graph, equipment and a storage medium. BACKGROUND

[0002] At present, in the CRM front-end handling process, the selection of a package mainly depends on artificial experience. When handling a business, a business personnel can only search for multiple related products that meet the requirements according to experience, and can only search for the products by a sales name or code, depending on the experience of the business personnel. The business handling efficiency is low, the product recommendation is prone to deviation, the recommendation effect is poor, it is difficult to support accurate marketing for customer demand, and the marketing effect is affected. SUMMARY

[0003] The embodiments of the present application provide a package recommendation method and device based on a knowledge graph, equipment and a storage medium to solve at least one problem in the related art. The technical solutions are as follows:

[0004] In a first aspect, the embodiments of the present application provide a package recommendation method based on a knowledge graph, comprising:

[0005] Obtaining input package information, and determining a first package according to the package information;

[0006] Determining at least one second package associated with the first package from the knowledge graph according to the first package;

[0007] Performing similarity analysis on the first package and the second package to determine a similarity result of the first package and the second package, wherein the similarity includes label similarity analysis and quantity similarity;

[0008] Determining a target package according to the similarity result, and recommending the target package.

[0009] In an implementation manner, the knowledge graph is constructed by the following steps:

[0010] Configuring entity attributes and relationship attributes in the knowledge graph through a configuration instruction;

[0011] Extracting entity information of a package corresponding to the entity attributes and relationship information of a package corresponding to the relationship attributes from a preset relational database;

[0012] Storing the entity attributes, the entity information, the relationship attributes and the relationship information in a graph database;

[0013] The entity attribute, the entity information, the relation attribute and the relation information are encapsulated as structured data to construct a knowledge graph.

[0014] In an embodiment, the determining at least one second package associated with the first package from the knowledge graph according to the first package comprises:

[0015] determining an associated product of the first package from the knowledge graph according to the first package;

[0016] determining at least one associated package having the same associated product from the knowledge graph according to the associated product;

[0017] performing traffic information analysis on the first package and at least one associated package to determine a traffic information analysis result;

[0018] determining at least one associated package with consistent traffic information analysis result as at least one second package associated with the first package.

[0019] In an embodiment, the performing traffic information analysis on the first package and at least one associated package to determine a traffic information analysis result comprises:

[0020] performing a first comparison on the number of traffic of the first package and the number of traffic of the associated package respectively, and determining a consistent traffic information analysis result when the first comparison result is the same;

[0021] or,

[0022] performing a second comparison on the type of traffic of the first package and the type of traffic of the associated package respectively, and determining a consistent traffic information analysis result when the second comparison result is the same.

[0023] In an embodiment, the similarity result comprises a label similarity analysis result; the label similarity analysis result is obtained by performing label similarity analysis on the first package and the second package, comprising:

[0024] determining a first label of the first package, a number of the first label, a second label of the second package and a number of the second label according to the knowledge graph, and determining a total number of labels according to the union set of the first label and the second label;

[0025] determining a same number of the second label which is the same as the first label;

[0026] determining a label similarity analysis result according to the ratio of the same number to the total number of labels.

[0027] In an implementation, the similarity result further comprises a quantity similarity analysis result; the quantity similarity analysis is performed on the first package and the second package to obtain the quantity similarity analysis result, which comprises:

[0028] According to the knowledge graph, a first cost value, a first traffic value, a first number of short messages, and a first voice duration of the first package, and a second cost value, a second traffic value, a second number of short messages, and a second voice duration of the second package are determined;

[0029] A first absolute value of a difference between the first cost value and the second cost value, a second absolute value of a difference between the first traffic value and the second traffic value, a third absolute value of a difference between the first number of short messages and the second number of short messages, and a fourth absolute value of a difference between the first voice duration and the second voice duration are determined;

[0030] A first sum of the first absolute value, the second absolute value, the third absolute value, and the fourth absolute value is determined;

[0031] According to a product of the first sum and a preset hyperparameter, a similarity distance is obtained, a second sum of the similarity distance and a preset numerical value is determined, and a ratio of the preset numerical value to the second sum is determined to obtain the quantity similarity analysis result.

[0032] In an implementation, the determining of the target package according to the similarity result comprises:

[0033] A third sum of the label similarity analysis result and the quantity similarity analysis result is determined, and a package similarity is determined according to half of the second sum;

[0034] The package similarities are arranged from high to low, and a preset number of second packages corresponding to the package similarities arranged in front are taken as target packages.

[0035] In a second aspect, an embodiment of the present application provides a package recommendation device based on a knowledge graph, comprising:

[0036] An acquisition module is configured to acquire input package information, and determine a first package according to the package information;

[0037] A determination module is configured to determine at least one second package associated with the first package from a knowledge graph according to the first package;

[0038] A similarity module is configured to perform similarity analysis on the first package and the second package, and determine a similarity result of the first package and the second package, wherein the similarity comprises label similarity analysis and quantity similarity;

[0039] The recommendation module is configured to determine a target package according to the similarity result, and recommend the target package.

[0040] In a third aspect, an electronic device is provided, which includes a processor and a memory. The memory stores instructions which are loaded and executed by the processor to implement the method in any of the embodiments of the above aspects.

[0041] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program. The computer program is executed to implement the method in any of the embodiments of the above aspects.

[0042] The beneficial effects of the above technical solutions at least include:

[0043] By obtaining the input package information, determining the first package according to the package information, and determining at least one second package associated with the first package from the knowledge graph according to the first package, the second package associated with the first package can be automatically determined based on the knowledge graph using the information of the first package, which is more efficient. The similarity analysis is performed on the first package and the second package to determine the similarity result of the first package and the second package, and the similarity includes label similarity analysis and quantity similarity, which improves the accuracy of the similarity analysis. According to the similarity result, the target package is determined and recommended, which reduces the dependence on human experience and is conducive to improving the recommendation effect and efficiency and facilitating the improvement of marketing effect.

[0044] The above summary is only for the purpose of the description and is not intended to limit in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present application will be readily apparent to those skilled in the art by reference to the drawings and the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0045] In the drawings, like reference numerals refer to same or similar functionalities throughout the several views. The drawings are not necessarily to scale. It is to be understood that the drawings only depict several embodiments of the disclosure and are not to be considered as limiting its scope.

[0046] Figure 1 A step flowchart of a package recommendation method based on a knowledge graph according to an embodiment of the present application;

[0047] Figure 2 A flowchart of a package recommendation method based on a knowledge graph according to an embodiment of the present application;

[0048] Figure 3 A structure block diagram of a package recommendation device based on a knowledge graph according to an embodiment of the present application;

[0049] Figure 4 Structure block diagram of electronic device of an embodiment of the present application. DETAILED DESCRIPTION

[0050] Hereinafter, only certain exemplary embodiments are simply described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present application. Therefore, the drawings and the description are considered to be exemplary in nature rather than limiting.

[0051] Glossary:

[0052] Graph Database: A data management system that takes point and edge as basic storage unit and takes efficient storage and query of graph data as design principle.

[0053] Knowledge Graph: In the library intelligence field, it is called knowledge domain visualization or knowledge field mapping map, which is a series of various different graphs showing the development process and structural relationship of knowledge, using visualization technology to describe knowledge resources and their carriers, mining, analyzing, constructing, drawing and displaying knowledge and their mutual relationship.

[0054] Offer: A tariff package set provided by a telecom operator, which contains voice call, Internet traffic and other resources.

[0055] Reference Figure 1 , a flowchart of a knowledge graph-based offer recommendation method of an embodiment of the present application is shown, which can at least include steps S100-S400:

[0056] S100, acquiring inputted offer information and determining a first offer according to the offer information.

[0057] S200, determining at least one second offer associated with the first offer from the knowledge graph according to the first offer.

[0058] S300, performing similarity analysis on the first offer and the second offer to determine the similarity result of the first offer and the second offer, and the similarity includes label similarity analysis and quantity similarity.

[0059] S400, determining a target offer according to the similarity result and recommending the target offer.

[0060] The knowledge graph-based offer recommendation method of the embodiment of the present application can be executed by the electronic control unit, controller, processor, etc. of the terminal such as computer, mobile phone, tablet, vehicle-mounted terminal, etc. or by the cloud server.

[0061] The technical scheme of the embodiment of the application can obtain input package information, determine a first package according to the package information, determine at least one second package associated with the first package from a knowledge graph according to the first package, automatically determine the associated second package based on the knowledge graph and the information of the first package, and be more efficient; perform similarity analysis on the first package and the second package, determine a similarity result of the first package and the second package, and improve the accuracy of the similarity analysis; determine a target package according to the similarity result, and recommend the target package, thereby reducing the dependence on human experience, improving the recommendation effect and efficiency, and facilitating the improvement of the marketing effect.

[0062] As shown in FIG. 1, in an embodiment, in step S100, a user (for example, a customer) can input package information on a terminal receiving system, for example, input a search condition such as a sales product name, search for a package containing a traffic volume > 20G, and then select a certain package, so as to determine the package information of the selected package, or when the customer consults a service on site, a staff member can input the package information on the terminal receiving system based on the customer's demand. Then, the system determines a first package corresponding to the current package information as a reference. Figure 2

[0063] In an embodiment, the knowledge graph is constructed by the following steps:

[0064] 1. Configure entity attributes and relationship attributes in the knowledge graph through a configuration instruction to obtain an ontology model. For example, the entity attributes include but are not limited to sales products (such as packages), products, tariffs, restrictions, tags, broadband, short messages, voice, and traffic volume; and the relationship attributes include but are not limited to associated products, mandatory, optional, default selection, containing, having a tariff, having a tag, and having a restriction. For example:

[0065] Sales product --[mandatory|optional|default selection]--> sales product

[0066] Sales product --[associated product]--> product

[0067] Sales product --[has tariff]--> tariff

[0068] Sales product --[has tag]--> tag (the tag type is not limited)

[0069] Sales product --[has restriction]--> restriction

[0070] Sales product --[contains]--> short message

[0071] Sales product --[contains]--> broadband

[0072] ​Sales product -- [contains] -- voice

[0073] Sales product -- [contains] -- traffic

[0074] 2. Then, according to the ontology model, the taking range of the entity attribute and the relationship attribute is confirmed, which refers to extracting the entity information of the package corresponding to the entity attribute and the relationship information of the package corresponding to the relationship attribute from the preset relational database. For example, the entity information of the sales product can be directly extracted from the sales product table, the relationship information of the sales product -- [associated product] -- product is extracted from the sales product associated product table, the relationship information of the sales product -- [contains] -- traffic can be extracted from the sales product tariff template instance table and the sales product attribute table, and the like.

[0075] 3. The entity attribute, the entity information, the relationship attribute and the relationship information are stored through the graph database neo4j. In addition, after the graph database neo4j is constructed, the structure of the graph data collection data can be further defined based on actual needs through the parameter configuration mode.

[0076] 4. The entity attribute, the entity information, the relationship attribute and the relationship information are encapsulated as structured data and synchronized to the knowledge graph platform to construct the knowledge graph. Then, after the sales product management department receives the demand to adjust the sales product configuration in the CPC configuration system, configuration auditing and publishing are performed, and the structure of the graph data collection data defined in the previous step is configured to update the knowledge graph in real time.

[0077] Wherein, when the attribute data of the sales product changes, the CPC system updates the sales product configuration data change information to the graph database according to the graph ontology model; the knowledge graph platform provides knowledge graph management and comprehensive query capabilities, the constructed knowledge graph is stored on the knowledge graph platform, and connection information of the knowledge graph is provided; the knowledge graph capability component is further provided in the embodiment of the application, the graph data is obtained through the connection of the knowledge graph platform, and multi-dimensional search and package similarity analysis capabilities are provided; the production business acceptance system such as CRM and new sales provides intelligent search service and similar sales product recommendation service for the acceptance personnel by calling the knowledge graph capability component, and the business handling efficiency is improved.

[0078] In the embodiment of the application, after the knowledge graph is constructed, multi-dimensional search of the sales product (taking the package as an example) can be realized. Since the entity attribute, the relationship attribute and the like of the sales product have been stored in the graph database, multi-dimensional search can be quickly performed according to the entity information and the relationship information of the sales product.

[0079] In one embodiment, as shown in FIG. 1, the system for constructing the knowledge graph of the sales product comprises a CPC configuration system, a knowledge graph platform and a production business acceptance system. Figure 2As shown, the similarity analysis component is introduced for retrieval, label similarity analysis and quantity similarity analysis; in an embodiment, step S200 includes steps S210-S240:

[0080] S210, according to the first package, determining the associated products of the first package from the knowledge graph.

[0081] Optionally, to perform product consistency judgment, according to the first package, performing retrieval in the knowledge graph, thereby determining which associated products of the first package are, for example, the associated products of the first package are [product C, product D].

[0082] S220, according to the associated products, determining at least one associated package with the same associated products from the knowledge graph.

[0083] Optionally, assuming that the associated products [product C, product D] are determined, at this time, [product C, product D] are used as a query statement, and retrieval is performed in the knowledge graph to see which packages have the same associated products [product C, product D], for example, nodes such as package B and package D are retrieved, and the associated products of package B and package D are both [product C, product D], at this time, package B and package D are both associated packages, that is, the nodes meet the conditions, and the similarity analysis component is returned, and package B and package D have product consistency with the first package. It should be noted that,

[0084] S230, performing traffic information analysis on the first package and at least one associated package to determine a traffic information analysis result.

[0085] Optionally, step S2301 or S2302 is included:

[0086] S2301, respectively comparing the traffic quantity of the first package with the traffic quantity of the associated packages, and when the first comparison result is the same, determining a traffic information analysis result representing consistency.

[0087] It should be noted that the traffic type can be ordinary traffic and directional traffic, and the traffic quantity refers to the number of traffic included in the package, for example, some packages can have 1 ordinary traffic and 1 directional traffic, at this time, the traffic quantity is 2, and some packages can have 1 ordinary traffic and 2 directional traffics, at this time, the traffic quantity is 3. Therefore, the traffic quantity of the first package is respectively compared with the traffic quantity of each associated package, when the first comparison result is the same, it is indicated that the traffic quantity of the first package is the same as the traffic quantity of at least one associated package, at this time, a traffic information analysis result representing consistency is determined. For example, if the traffic quantity of the first package and the associated package B is both 2, at this time, the traffic information analysis result of the first package and the associated package B represents consistency.

[0088] S2302, respectively compare the traffic types of the first package with the traffic types of the associated packages, and when the second comparison result is the same, determine the traffic information analysis result that is consistent in representation.

[0089] In another way, the traffic types of the first package can be respectively compared with the traffic types of the associated packages, and when the second comparison result is the same, the traffic information analysis result that is consistent in representation is determined. For example, if the traffic types of the first package and the associated package B both have directional traffic and ordinary traffic, the traffic information analysis results of the first package and the associated package B are consistent in representation at this time.

[0090] S240, at least one associated package with consistent traffic information analysis result is determined as at least one second package associated with the first package.

[0091] Then, at least one associated package with consistent traffic information analysis result determined by one of the above ways is determined as at least one second package associated with the first package.

[0092] In an embodiment, as shown in Figure 2 If there is no node that meets the conditions, there is no second package at this time, and the query result is returned and displayed; and if there is a second package that meets the conditions, the similarity analysis component is used to perform label similarity analysis and quantity similarity analysis, and the similarity is calculated, so that the similarity result includes the label similarity analysis result and the quantity similarity analysis result; the first package and the second package are subjected to label similarity analysis in step S300, and the label similarity analysis result is obtained, including steps S310-S330:

[0093] S310, according to the knowledge graph, determining the first label of the first package, the number of the first label, the second label of the second package, and the number of the second label, and determining the total number of labels according to the union set of the first label and the second label.

[0094] Optionally, according to the knowledge graph, the first label (for example, there are label 1, label 2, and label 3) of the first package, the number of the first label (which is 3 at this time), the second label (for example, there are label 2 and label 4) of the second package, and the number of the second label (which is 2 at this time) are determined, and the union set of the first label and the second label is determined: (label 1, label 2, label 3, and label 4), so that the total number of labels is determined to be 4.

[0095] S320, determining the same number of the second label that is the same as the first label.

[0096] For example, the first label has label 1, label 2, and label 3, and the second label has label 2 and label 4, and at this time, label 2 is the same, so the same number is 1.

[0097] S330, determining the label similarity analysis result according to the ratio of the same number to the sum of the label numbers.

[0098] Specifically, as in the above example, the label similarity analysis result

[0099] In an embodiment, the step S300 of performing the quantity similarity analysis on the first package and the second package to obtain the quantity similarity analysis result includes steps S340-S370:

[0100] S340, determining a first fee value, a first traffic value, a first message number, and a first voice duration of the first package, and a second fee value, a second traffic value, a second message number, and a second voice duration of the second package according to the knowledge graph.

[0101] Optionally, taking the first package as package A and the second package as package B as an example, the first fee value fee A , the first traffic value traffic A , the first message number message A , and the first voice duration voice A of the first package, and the second fee value fee B , the second traffic value traffic B , the second message number message B , and the second voice duration voice B of the second package are determined according to the knowledge graph.

[0102] S350, determining a first absolute value of a difference between the first fee value and the second fee value, a second absolute value of a difference between the first traffic value and the second traffic value, a third absolute value of a difference between the first message number and the second message number, and a fourth absolute value of a difference between the first voice duration and the second voice duration.

[0103] Optionally, a first absolute value |fee A -fee B | of the difference between the first fee value and the second fee value, a second absolute value |traffic A -traffic B | of the difference between the first traffic value and the second traffic value, a third absolute value |message A -message B | of the difference between the first message number and the second message number, and a fourth absolute value |voice A -voice B | of the difference between the first voice duration and the second voice duration are determined.

[0104] S360, determining a first sum value of the first absolute value, the second absolute value, the third absolute value and the fourth absolute value.

[0105] Optionally, the first sum value of the first absolute value, the second absolute value, the third absolute value and the fourth absolute value is determined as follows: |fee A -fee B |+|traffic A -traffic B |+|message A -message B |

[0106] +|voice A -voice B |.

[0107] S370, obtaining a similarity distance according to a product of the first sum value and a preset hyperparameter, determining a second sum value of the similarity distance and a preset numerical value, and determining a ratio of the preset numerical value and the second sum value to obtain a quantitative similarity analysis result.

[0108] It should be noted that the preset hyperparameter a can be set to 0.01 to prevent the difference from being too large or too small to affect the similarity calculation, and can be set as needed. Then, according to the product of the first sum value |fee A -fee B |+|traffic A -traffic B |+|message A -message B |+|voice A -voice B | and the preset hyperparameter a, the similarity distance distance is obtained as follows:

[0109] distance = a * (|fee A -fee B |+|traffic A -traffic B |+|message A -message B |

[0110] |+|voice A -voice B |)

[0111] Optionally, taking the preset value 1 as an example, a second sum (1+distance) of the similarity distance distance and the preset value 1 is determined, and a ratio of the preset value 1 to the second sum (1+distance) is determined to obtain a quantity similarity analysis result sim num :

[0112]

[0113] In an embodiment, the determining the target package according to the similarity result in step S400 includes steps S410-S420:

[0114] S410, determining a third sum of the label similarity analysis result and the quantity similarity analysis result, and determining a package similarity according to half of the third sum.

[0115] Optionally, the third sum of the label similarity analysis result sim label and the quantity similarity analysis result sim num is (sim label +sim num ).

[0116] Optionally, the calculation formula of the package similarity sim out is:

[0117]

[0118] S420, arranging the package similarities from high to low, and taking the preset number of package similarities corresponding to the second packages arranged at the front (i.e., the second packages) as the target packages.

[0119] Optionally, for example, there are five second packages B1, B2, B3, B4 and B5, and the package similarities sim out calculated respectively are 0.5, 0.6, 0.4, 0.3 and 0.2. The package similarities are arranged from high to low as 0.6, 0.5, 0.4, 0.3 and 0.2, and the corresponding second packages are the second package B2, the second package B1, the second package B3, the second package B4 and the second package B5. Assuming that the preset number is 3, the second packages corresponding to the three package similarities arranged at the front are determined as the target packages, i.e., the second package B2, the second package B1 and the second package B3 are taken as the target packages. Finally, the target packages are recommended, for example, displayed in a terminal.

[0120] The method of the embodiment of the application constructs the analysis and recommendation capability of similar packages, can calculate other packages similar to the customer demand based on attribute information such as entity attributes and relationship attributes of the package, compared with the traditional relational database, the similarity needs to be calculated one by one for all sales products across multiple tables, the process is slow and cannot be practically used in production, the sales product similarity algorithm defined in the embodiment of the application is directly calculated based on the data of the sales product graph, and a batch of sales products that do not meet the precondition are filtered in advance so as to reduce the calculation data and improve the efficiency; the search provides more diversified selection items, the functions of the package intelligent search, similarity analysis and recommendation developed based on the knowledge graph support the multi-dimensional search service and the similar package recommendation function of the package during the business acceptance, the multi-dimensional search of the sales product is enabled by the efficient search of the graph database, and the defined multi-dimensional search mode takes into account the generality and flexibility, the node and relationship query factors are summarized and refined in the consideration of the general search of the graph nodes and relationships, so that the node or relationship can be used for personalized query in a more flexible way, and the handling efficiency of the marketing business acceptance personnel is improved.

[0121] At the same time, since it is a private professional package knowledge graph, the transmission, storage and use are internal, the safety of the key data is ensured, and the data of the knowledge graph is synchronized to the knowledge graph in real time through a certain strategy, so that the data of the knowledge graph is ensured to be in the latest state; the knowledge graph is applied to the field of the telecommunications industry, the private sales product knowledge graph is constructed by using the data advantage of the sales product centralized management system, the information data safety is protected, and the utilization rate and development direction of the management system for the data are improved and widened, which provides a new idea and scheme for subsequent data processing and use.

[0122] With reference to Figure 3 , a structure block diagram of a package recommendation device based on a knowledge graph is shown, the device can include:

[0123] The acquisition module is configured to acquire the input package information, and determine a first package according to the package information;

[0124] The determination module is configured to determine at least one second package associated with the first package from the knowledge graph according to the first package;

[0125] The similarity module is configured to perform similarity analysis on the first package and the second package, and determine the similarity result of the first package and the second package, the similarity includes label similarity analysis and quantity similarity;

[0126] The recommendation module is configured to determine a target package according to the similarity result, and recommend the target package.

[0127] The functions of each module in each device of the embodiments of the present application can be referred to the corresponding description in the above method, which will not be repeated here.

[0128] Referring to Figure 4 , a structural block diagram of an electronic device according to an embodiment of the present application is shown, which includes a memory 310 and a processor 320, the memory 310 stores instructions executable on the processor 320, and the processor 320 loads and executes the instructions to implement the knowledge graph-based package recommendation method in the above embodiments. The number of the memory 310 and the processor 320 can be one or more.

[0129] In an implementation, the electronic device further includes a communication interface 330 for communicating with external devices and performing data interaction transmission. If the memory 310, the processor 320 and the communication interface 330 are independently implemented, the memory 310, the processor 320 and the communication interface 330 can be connected to each other through a bus and complete communication therebetween. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 In the figure, only one thick line is used to represent, but it does not mean that there is only one bus or only one type of bus.

[0130] Optionally, in specific implementation, if the memory 310, the processor 320 and the communication interface 330 are integrated on a chip, the memory 310, the processor 320 and the communication interface 330 can complete communication therebetween through an internal interface.

[0131] The embodiments of the present application provide a computer readable storage medium, which stores a computer program, the computer program is executed by a processor to implement the knowledge graph-based package recommendation method provided in the above embodiments.

[0132] The embodiments of the present application further provide a chip, which includes a processor, and is used for calling and running instructions stored in a memory, so that a communication device installed with the chip executes the method provided in the embodiments of the present application.

[0133] The embodiment of the present application further provides a chip, comprising: an input interface, an output interface, a processor and a memory, the input interface, the output interface, the processor and the memory are connected through internal connection paths, the processor is used for executing the code in the memory, and when the code is executed, the processor is used for executing the method provided by the embodiment of the present application.

[0134] It should be understood that the processor described above can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. It should be noted that the processor can be a processor supporting an advanced RISC machine (ARM) architecture.

[0135] Further, the aforementioned memory can include a read-only memory, and a random access memory, and can further include a nonvolatile random access memory. The memory can be a volatile memory or a nonvolatile memory, or can include both volatile and nonvolatile memory. The nonvolatile memory can include a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), or a flash memory. The volatile memory can include a random access memory (RAM), which functions as an external cache. Many forms of RAM are available. For example, a static RAM (SRAM), a dynamic RAM (DRAM), a synchronous DRAM (SDRAM), a double data rate SDRAM (DDR SDRAM), an enhanced SDRAM (ESDRAM), a synchlink DRAM (SLDRAM), and a direct rambus RAM (DR RAM) are available.

[0136] In the above-described embodiments, all or a part can be implemented by software, hardware, firmware, or any combination thereof. When implemented as software, it can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed by a computer, all or a part of the procedures or functions according to the present disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium.

[0137] In the description of the application, the description of the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the application. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in one or more embodiments or examples. In addition, different embodiments or examples described in the specification and the features of different embodiments or examples can be combined and combined by those skilled in the art without contradiction, if necessary.

[0138] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0139] Any process or method descriptions in flow charts or described elsewhere herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for implementing specific logic functions or other processes. And the various embodiments of the application can include additional or fewer steps or processes in comparison to those shown in the figures.

[0140] The logic and / or steps represented in flow charts or otherwise described herein, for example, can be embodied in computer-readable instructions, which can be used to cause one or more processors to perform the actions indicated in the steps. The computer-readable instructions can be stored on one or more storage media or memory devices associated with the one or more processors.

[0141] It should be understood that parts of the application can be implemented in hardware, software, firmware or a combination thereof. In the above-described embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above-described embodiment method can be instructed by a program to complete the relevant hardware, which can be stored in a computer-readable storage medium, and the program includes one or a combination of the steps of the method embodiment when executed.

[0142] In addition, each of the function units in each embodiment of the present application can be integrated in one processing module, or each unit can be physically present separately, or two or more units can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software function module. When the integrated module is realized in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium. The storage medium can be a read-only memory, a magnetic disk or an optical disk, etc.

[0143] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of various changes or replacements within the technical scope disclosed in the present application, and these should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A knowledge graph-based package recommendation method, characterized in that, The method comprises the following steps: acquiring input package information and determining a first package according to the package information; determining at least one second package associated with the first package from a knowledge graph according to the first package; performing similarity analysis on the first package and the second package to determine a similarity result of the first package and the second package, the similarity comprising label similarity analysis and quantity similarity; determining a target package according to the similarity result and recommending the target package. 2.The knowledge graph-based package recommendation method of claim 1, wherein: The knowledge graph is constructed by the following steps: configuring entity attributes and relationship attributes in the knowledge graph through a configuration instruction; extracting entity information of a package corresponding to the entity attributes and relationship information of a package corresponding to the relationship attributes from a preset relational database; storing the entity attributes, the entity information, the relationship attributes and the relationship information in a graph database; encapsulating the entity attributes, the entity information, the relationship attributes and the relationship information as structured data to construct the knowledge graph. 3.The knowledge graph-based package recommendation method of claim 1, characterized in that: The step of determining at least one second package associated with the first package from the knowledge graph according to the first package comprises the following steps: determining an associated product of the first package from the knowledge graph according to the first package; determining at least one associated package having the same associated product from the knowledge graph according to the associated product; performing traffic information analysis on the first package and at least one associated package to determine a traffic information analysis result; determining at least one associated package with consistent traffic information analysis result as at least one second package associated with the first package. 4.The knowledge graph-based package recommendation method of claim 3, characterized in that: The step of performing traffic information analysis on the first package and at least one associated package to determine a traffic information analysis result comprises the following steps: performing a first comparison between the traffic quantity of the first package and the traffic quantity of the associated package, and determining a consistent traffic information analysis result when the first comparison result is the same; or performing a second comparison between the traffic type of the first package and the traffic type of the associated package, and determining a consistent traffic information analysis result when the second comparison result is the same. 5.The knowledge graph-based package recommendation method according to any one of claims 1-4, characterized in that: The similarity result comprises a label similarity analysis result. The step of performing label similarity analysis on the first package and the second package to obtain a label similarity analysis result comprises the following steps: determining a first label of the first package, a quantity of the first label, a second label of the second package and a quantity of the second label according to the knowledge graph, and determining a total quantity of labels according to the union of the first label and the second label; determining a same quantity of the second label that is the same as the first label; determining a label similarity analysis result according to the ratio of the same quantity to the total quantity of labels. 6.The method of claim 5, wherein: The similarity result further comprises a quantity similarity analysis result. The step of performing quantity similarity analysis on the first package and the second package to obtain a quantity similarity analysis result comprises the following steps: According to the knowledge graph, a first cost value, a first traffic value, a first number of short messages, and a first voice duration of the first package and a second cost value, a second traffic value, a second number of short messages, and a second voice duration of the second package are determined; A first absolute value of a difference between the first cost value and the second cost value, a second absolute value of a difference between the first traffic value and the second traffic value, a third absolute value of a difference between the first number of short messages and the second number of short messages, and a fourth absolute value of a difference between the first voice duration and the second voice duration are determined; A first sum value of the first absolute value, the second absolute value, the third absolute value, and the fourth absolute value is determined; According to a product of the first sum value and a preset hyperparameter, a similarity distance is obtained, a second sum value of the similarity distance and a preset numerical value is determined, and a ratio of the preset numerical value and the second sum value is obtained to obtain a quantity similarity analysis result. 7.The knowledge graph-based package recommendation method of claim 6, characterized in that: The similarity result includes: A third sum value of the label similarity analysis result and the quantity similarity analysis result is determined, a package similarity is determined according to half of the second sum value; The package similarity is arranged from high to low, and a preset number of second packages corresponding to the package similarity in front of the arrangement are taken as target packages. 8.A knowledge graph-based package recommendation apparatus, characterized by comprising: The method includes: An acquisition module is configured to acquire input package information and determine a first package according to the package information; A determination module is configured to determine at least one second package associated with the first package from a knowledge graph according to the first package; A similarity module is configured to perform similarity analysis on the first package and the second package to determine a similarity result of the first package and the second package, the similarity including label similarity analysis and quantity similarity; A recommendation module is configured to determine target packages according to the similarity result and recommend the target packages.

9. An electronic device, comprising: The method includes: A processor and a memory, the memory stores instructions, the instructions are loaded and executed by the processor to implement the method of any one of claims 1-7.

10. A computer readable storage medium, the computer readable storage medium stores a computer program, the computer program is executed to implement the method of any one of claims 1-7.

Images