CN116167635A

CN116167635A - Method and device for improving evaluation accuracy

Info

Publication number: CN116167635A
Application number: CN202211516798.6A
Authority: CN
Inventors: 陈映雪
Original assignee: China Construction Bank Corp; CCB Finetech Co Ltd
Current assignee: China Construction Bank Corp; CCB Finetech Co Ltd
Priority date: 2022-11-29
Filing date: 2022-11-29
Publication date: 2023-05-26

Abstract

The embodiment of the invention relates to the technical field of data processing, and discloses a method and a device for improving evaluation accuracy, wherein an evaluation field of a target object is acquired from an external data interface; cleaning the evaluation fields, and classifying the cleaned evaluation fields according to preset scene dimensions to obtain classification results; generating a relationship map from the evaluation field by using a graph database based on the classification result; and processing the relation map by using a random forest algorithm to obtain a field set with the evaluation influence degree of the target object being larger than the preset influence degree. The method and the device solve the technical problem that in the prior art, evaluation under different scenes exists when single scene evaluation is carried out on the target object, and the evaluation under different scenes has overlapping or complementation, make up the defect of evaluation data of the target object, and achieve the beneficial effect of improving the evaluation accuracy of the target object.

Description

Method and device for improving evaluation accuracy

Technical Field

The embodiment of the invention relates to the technical field of data processing, in particular to a method and a device for improving evaluation accuracy.

Background

The score of the target object is loose currently, and different scenes have different evaluation results on the target object. However, the complete evaluation condition of the target object cannot be reflected by the evaluation of a single scene, and the evaluation of the target object may be overlapped or complemented under different scenes, so that the current technology cannot find the overlapped or complemented condition, and the use efficiency of the evaluation result of the target object is low.

Disclosure of Invention

The embodiment of the invention provides a method and a device for improving evaluation accuracy, which solve the technical problem that in the prior art, evaluation under different scenes has overlapping or complementation when single scene evaluation is carried out on a target object.

According to an aspect of the present invention, there is provided a method for improving accuracy of evaluation, the method comprising:

the method comprises the steps of obtaining evaluation fields of a target object through an external data interface, wherein the number of the external data interfaces is multiple, the number of the evaluation fields is multiple, one external data interface correspondingly has multiple evaluation fields, and external data refer to all data obtained through the external data interface except known data of the external data interface;

cleaning the evaluation field, and classifying the cleaned evaluation field according to a preset scene dimension to obtain a classification result, wherein the preset scene dimension is obtained according to scene division required by evaluating the target object;

generating a relationship graph from the evaluation field by using a graph database based on the classification result;

and processing the relation map by using a random forest algorithm to obtain a field set with the evaluation influence degree of the target object being larger than a preset influence degree.

According to another aspect of the present invention, there is provided an apparatus for improving accuracy of evaluation, the apparatus comprising:

a field obtaining unit, configured to obtain, through an external data interface, an evaluation field of a target object, where the number of external data interfaces is multiple, the number of evaluation fields is multiple, one external data interface corresponds to the existence of multiple evaluation fields, and external data refers to all data obtained through the external data interface except for self known data;

the field processing unit is used for cleaning the evaluation field and classifying the cleaned evaluation field according to a preset scene dimension to obtain a classification result, wherein the preset scene dimension is obtained according to scene division required by evaluating the target object;

a map generation unit for generating a relationship map using a graphic database from the evaluation field based on the classification result;

and the evaluation field determining unit is used for processing the relation map by using a random forest algorithm to obtain a field set with the evaluation influence degree of the target object being larger than a preset influence degree.

According to another aspect of the present invention, there is provided an apparatus for improving accuracy of evaluation, the apparatus for improving accuracy of evaluation including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method for improving accuracy of assessment according to any one of the embodiments of the present invention.

According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute a method for improving accuracy of evaluation according to any one of the embodiments of the present invention.

The embodiment of the invention discloses a method and a device for improving evaluation accuracy, which are used for acquiring an evaluation field of a target object from an external data interface; cleaning the evaluation fields, and classifying the cleaned evaluation fields according to preset scene dimensions to obtain classification results; generating a relationship map from the evaluation field by using a graph database based on the classification result; and processing the relation map by using a random forest algorithm to obtain a field set with the evaluation influence degree of the target object being larger than the preset influence degree. The method and the device solve the technical problem that in the prior art, evaluation under different scenes exists when single scene evaluation is carried out on the target object, and the evaluation under different scenes has overlapping or complementation, make up the defect of evaluation data of the target object, and achieve the beneficial effect of improving the evaluation accuracy of the target object.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of a method for improving evaluation accuracy according to an embodiment of the present invention;

FIG. 2 is a block diagram of a relationship graph of dimension-interface-fields provided by an embodiment of the present invention;

FIG. 3 is a block diagram of a device for improving evaluation accuracy according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of an apparatus for improving accuracy of evaluation according to an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

It should be noted that the terms "first," "second," "target," and the like in the description and claims of the present invention and in the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Fig. 1 is a flowchart of a method for improving evaluation accuracy, which is provided in an embodiment of the present invention, and the embodiment may be suitable for situations such as evaluating effects of products or performing credit evaluation on clients, where the method may be performed by an apparatus for improving evaluation accuracy, and the apparatus for improving evaluation accuracy may be implemented in a form of hardware and/or software, and may be generally integrated in a server. The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations.

As shown in fig. 1, the method for improving the evaluation accuracy specifically includes the following steps:

s101, acquiring evaluation fields of a target object through external data interfaces, wherein the number of the external data interfaces is multiple, the number of the evaluation fields is multiple, one external data interface correspondingly has multiple evaluation fields, and external data refers to all other data acquired through the external data interfaces except for self known data.

Illustratively, for the example of an enterprise, the external data is typically data provided by organizations and institutions outside of the enterprise's body, and the provider of these external data typically provides an external data interface to the enterprise to enable the enterprise to obtain the corresponding external data as needed. Because the number of external data providers and external data interfaces is increased, the external data provided by different external data providers may belong to different scenes or data fields, and external data acquired by different external data interfaces need to be complemented, so that the acquired evaluation data of the target object is more complete, and the evaluation of the target object is more complete.

Specifically, the external data acquired through the external data interface is an evaluation field of a target object, and the target object is not a single product, person or item, but a product, crowd or item. Assuming that the target object is a student in a bank customer, the target object needs to be evaluated, and about 95 external data interfaces capable of acquiring the target object, including credit points, repayment capability and the like of the customer, have about 2000 evaluation fields.

S102, cleaning the evaluation field, and classifying the cleaned evaluation field according to a preset scene dimension to obtain a classification result, wherein the preset scene dimension is obtained according to scene division required by evaluating the target object.

Specifically, when there are 2000 evaluation fields, the meanings represented by some of the evaluation fields such as a customer name, a customer identity, etc. are similar in nature, so that it is necessary to perform a cleaning process on the acquired evaluation fields, including merging of the similar fields, and deriving the current meanings of some of the evaluation fields. Illustratively, some evaluation fields such as 7 days, 30 days or 60 days of the overdue of the credit card represent overdue of the overdue, but the overdue times and durations represent different repayment capacities of the customers, so that the calculated overdue times and the duty ratio of the overdue durations of the customers are needed to derive new evaluation fields. After the merging process, only 1832 fields are left in 2000 evaluation fields, and after the deriving process, 1832 fields are raised to 2215 fields again.

After the evaluation fields are cleaned, the obtained 2215 evaluation fields are classified according to preset scene dimensions, for example, when the target object is a student in a banking customer, the preset scene dimensions can include five parts of credit, risk level, lending intention level, repayment capability level and stability level, after the classification according to the preset scene dimensions, as one external data interface can appear in multiple dimensions, since the content of one external data interface possibly hits keywords of two preset scene dimensions at the same time, for example, some external data interfaces name risk interfaces, but the data content of the interface can contain indexes such as repayment capability and liability capability of the target object, and the indexes can affect both the stability and the lending capability at the same time, therefore, after the classification is simple, the relationship between different evaluation fields and each preset scene dimension and the relatedness between the two evaluation fields are also required to be judged, so that a final classification result is obtained.

And S103, generating a relation graph by using a graph database from the evaluation field based on the classification result.

Specifically, after classifying the evaluation fields, a relationship graph is generated by using a graph database based on the relationship between the evaluation fields and preset scenario dimensions and external data interfaces, fig. 2 is a block diagram of a dimension-interface-field relationship graph provided in the embodiment of the present invention, referring to fig. 2, if a bank evaluates student clients in clients, the preset scenario dimensions include five parts of credit, risk level, loan intention level, repayment capability level, and stability level, and each preset scenario dimension corresponds to some of the 95 external data interfaces, such as a fraud portrait interface, a personal credit interface, a personal repayment capability interface, a loan intention interface, a annual income interface, and a repayment stability interface shown in fig. 2, where the evaluation field corresponding to the preset scenario dimension under each external data interface is also displayed in the relationship graph. The relation map can clearly display the relation among the evaluation field, the preset scene dimension and the external data interface.

It should be noted that, the relationship affinity between the evaluation field and the evaluation field will also generate another field 1-field 2-affinity relationship map by using the graphic database, which is similar to the relationship map of the dimension-interface-field shown in fig. 2, and will not be described herein.

And S104, processing the relation map by using a random forest algorithm to obtain a field set with the evaluation influence degree of the target object being larger than the preset influence degree.

Specifically, a random forest is a classifier that contains multiple decision trees, and whose output class is a mode of the class output by the individual trees. After a forest is established in a random manner, the forest is composed of a plurality of decision trees, and each decision tree of the random forest is irrelevant. After a forest is obtained, when a new input sample enters, each decision tree in the forest is judged to see which type the sample belongs to, and then which type is selected most, and the sample is predicted to be which type. The random forest can process the quantity with the attribute as a discrete value or the quantity with the attribute as a continuous value, and can be used for unsupervised learning clustering and abnormal point detection.

After the dimension-interface-field relation map and the field 1-field 2-intimacy relation map are obtained, the two relation maps are processed by using a random forest algorithm, fields with higher intimacy are removed, and finally a field set with evaluation influence on the target object being larger than preset influence is obtained, namely the field set with higher evaluation influence on the target object is obtained for later evaluation.

The method and the device solve the technical problem that in the prior art, evaluation under different scenes exists when single scene evaluation is carried out on the target object, and the evaluation under different scenes has overlapping or complementation, make up the defect of evaluation data of the target object, and achieve the beneficial effect of improving the evaluation accuracy of the target object.

Based on the above technical solutions, S102, performing cleaning treatment on the evaluation field specifically includes: combining similar field processing is carried out on the evaluation fields to obtain a primary processing result; and carrying out derivative field processing on the primary processing result to obtain a secondary processing result.

Specifically, since the evaluation of the target object may have overlapping portions or complementary portions in different scenes, in order to ensure that the evaluation of the target object is more comprehensive and complete, the existence of repeated or overlapping evaluation is avoided, and after the evaluation field of the target object is obtained, cleaning treatment is required to be performed on the evaluation field. Specifically, the evaluation fields need to be combined, that is, similar fields representing similar meanings in the evaluation fields are combined, for example, the name of the target object and the model of the target object each represent the identity of the target object, and then the same can be combined, and after the similar fields are combined, a primary processing result of the evaluation fields is obtained.

After the merging process, some evaluation fields have derived meanings, for example, the warranty period of the target object is one month, three months or one year, although the target object has the warranty period, different time periods of the warranty period bring different experiences to users, so that the evaluation of the target object can be different, and the corresponding relation between the time period of the warranty period and the after-sale times of the target object can be determined according to the time period of the warranty period, so that a new evaluation field is derived, and the evaluation of the target object is more complete and comprehensive. And after the evaluation field in the primary processing result is subjected to derivative field processing, obtaining a secondary processing result.

Based on the above technical solutions, S102, classifying the cleaned evaluation field according to a preset scene dimension, where the obtaining a classification result specifically includes: judging the relation between the evaluation field in the external data interface and each dimension in the preset scene dimensions, and storing the judging result into a first database; and judging the relation between the evaluation fields, and storing the judging result into a second database.

Specifically, after the evaluation field is cleaned, the cleaned evaluation field is further classified according to a preset scene dimension, and the classification result is stored in a corresponding database. In particular, since one external data interface can acquire a plurality of evaluation fields, a plurality of preset scene dimensions may be corresponding to one evaluation field, i.e., one external data interface may appear in a plurality of preset scene dimensions.

Illustratively, when the target object is a student in a banking customer, the preset scene dimension may include five parts of credit score, risk level, lending intention level, repayment capability level, stability level; in the corresponding evaluation fields, the evaluation fields belonging to the "risk level" dimension may include: the certificate number risk index, the under-credit risk index and the like, and the certificate number risk index and the under-credit risk index can also represent repayment capacity, thus also belonging to the dimension of repayment capacity grade.

Judging the relationship affinity between the cleaned evaluation fields and each preset scene dimension, storing the judging result in a first database, and meanwhile, judging the relationship affinity between different evaluation fields, for example, the more the number of preset scene dimensions in which two different evaluation fields fall simultaneously, the more intimate the relationship between the two evaluation fields, accordingly, judging the relationship between the evaluation fields and storing the judging result in a second database.

Based on the above technical solutions, determining a relationship between an evaluation field in the external data interface and each dimension in the preset scene dimensions, and storing the determination result in the first database includes: judging whether an evaluation field in an external data interface falls into one or more dimensions in a preset scene; if the evaluation field falls into one or more dimensions, the relation value between the evaluation field and the corresponding dimension is recorded as 1; if the evaluation field does not fall into any dimension, the relation value between the evaluation field and the corresponding dimension is recorded as 0; and storing the evaluation field and the preset scene dimension into a first database according to the triplet of the dimension-field-relation value.

Optionally, before determining the relationship between the evaluation field in the external data interface and each dimension in the preset scene dimensions, the method for improving the evaluation accuracy further includes: and performing word segmentation on the evaluation field by using an ES word segmentation device.

Specifically, the ES word segmentation device is used to convert a text into a series of words, i.e. a word segmentation process, for example, the ES word segmentation device segments keywords of related evaluation fields belonging to repayment capability, so as to obtain how many evaluation fields exist in a preset scene dimension belong to the dimension. After word segmentation, judging whether an evaluation field acquired from an external data interface falls into one or more dimensions of a preset scene; if the evaluation field hits the dimension, the relation value between the dimensions of the evaluation field is recorded as 1; if the evaluation field does not hit a certain dimension, the relation value between the evaluation field and the dimension is recorded as 0; and then storing the evaluation field and the preset scene dimension into a first database according to the triplet relation of the dimension-field-relation value.

Based on the above technical solutions, determining the relationship between the evaluation fields, and storing the determination result in the second database includes: judging whether the first evaluation field and the second evaluation field are obtained through the same external data interface, wherein the first evaluation field and the second evaluation field are any two fields in the evaluation fields; if yes, the relation value between the first evaluation field and the second evaluation field is marked as 0, and if not, the relation value between the first evaluation field and the second evaluation field is marked as 1; judging the number of dimensions hit by the first evaluation field and the second evaluation field together, and adding 1 to the relation value between the first evaluation field and the second evaluation field when one dimension hit by each time; and storing the relation value between the evaluation fields into a second database according to the triples of the field 1-field 2-relation values.

Specifically, since the evaluation fields are also related, it is necessary to determine whether any two of the evaluation fields, that is, the first evaluation field and the second evaluation field, are obtained through the same external data interface, and since the fields of the same external data interface do not have the same business meaning, if the first evaluation field and the second evaluation field are obtained through the same external data interface, the relationship value is recorded as 0, otherwise, the relationship value is recorded as 1; the first and second rating fields are then further evaluated (assuming a total of five preset scene dimensions):

whether the same preset scene dimension is hit or not, if so, the relation value is recorded as 1, otherwise, the relation value is recorded as 0;

whether the two preset scene dimensions are hit or not, if so, the relation value is recorded as 1, otherwise, the relation value is recorded as 0;

if hit with three preset scene dimensions, if yes, the relation value is recorded as 1, otherwise, the relation value is recorded as 0;

whether the four preset scene dimensions are hit or not, if so, the relation value is recorded as 1, otherwise, the relation value is recorded as 0;

if hit with five preset scene dimensions, if yes, the relation value is recorded as 1, otherwise, the relation value is recorded as 0;

accumulating the relation values to obtain field affinity between the first evaluation field and the second evaluation field, taking the field affinity as the relation value between the two evaluation fields, and storing the field affinity into a second database according to the triples of the field 1-field 2-relation values.

It should be noted that, in order to determine the relationship between the two evaluation fields more accurately, it may also be determined whether the first evaluation field and the second evaluation field are applied to the same service, if yes, the relationship value is recorded as 1, otherwise, the relationship value is recorded as 0, and the relationship value is accumulated into the relationship value as the relationship value between the two evaluation fields.

Based on the above technical solutions, S103, generating a relationship graph using a graph database for the evaluation field based on the classification result includes: based on the stored classification results in the first database and the second database, a relational graph of the evaluation field is generated by utilizing the Neo4j graph database.

Specifically, neo4j is a high-performance, noSQL (non-relational database) database that stores structured data on a network rather than in tables. It is an embedded disk-based java persistence engine with full transaction characteristics. Neo4j can also be viewed as a high performance graph engine with mature database ownership. Neo4j can build a knowledge graph based on the data type and the powerful searching function of the graph by the relation among knowledge points, and helps a user to search the associated knowledge. Based on the above function of Neo4j, the classified evaluation fields stored in the first database and the second database may generate a corresponding knowledge graph, that is, the above relationship graph. See fig. 2.

Based on the above technical solutions, after obtaining the field set with the evaluation influence on the target object greater than the preset influence in S104, the method for improving the evaluation accuracy further includes: and when the object to be evaluated belongs to the target object, evaluating the object to be evaluated by using the field set.

Specifically, because the target object represents a product, crowd or item, the obtained field set is also used for evaluating an object, after the field set is obtained, when the object to be evaluated needs to be evaluated, if the object to be evaluated belongs to the target object, the field set of the target object can be used for evaluating the object to be evaluated, and the effect of rapidly and accurately obtaining the complete evaluation result can be achieved.

Fig. 3 is a block diagram of a device for improving accuracy of evaluation provided in an embodiment of the present invention, where, as shown in fig. 3, the device for improving accuracy of evaluation includes:

a field obtaining unit 31, configured to obtain, through an external data interface, an evaluation field of the target object, where the number of external data interfaces is multiple, the number of evaluation fields is multiple, and one external data interface correspondingly has multiple evaluation fields, and external data refers to all data obtained through the external data interface except for self known data;

the field processing unit 32 is configured to perform cleaning processing on the evaluation field, and classify the cleaned evaluation field according to a preset scene dimension, to obtain a classification result, where the preset scene dimension is obtained according to a scene division required for evaluating the target object;

a map generation unit 33 for generating a relationship map using the graph database for the evaluation field based on the classification result;

the evaluation field determining unit 34 is configured to process the relationship map by using a random forest algorithm, so as to obtain a field set with an evaluation influence on the target object greater than a preset influence.

Optionally, the field processing unit 32 is further configured to:

combining similar field processing is carried out on the evaluation fields to obtain a primary processing result;

and carrying out derivative field processing on the primary processing result to obtain a secondary processing result.

Optionally, the field processing unit 32 is further configured to:

judging the relation between the evaluation field in the external data interface and each dimension in the preset scene dimensions, and storing the judging result into a first database;

and judging the relation between the evaluation fields, and storing the judging result into a second database.

Optionally, the field processing unit 32 is specifically configured to:

judging whether an evaluation field in an external data interface falls into one or more dimensions in a preset scene;

if the evaluation field falls into one or more dimensions, the relation value between the evaluation field and the corresponding dimension is recorded as 1;

if the evaluation field does not fall into any dimension, the relation value between the evaluation field and the corresponding dimension is recorded as 0;

and storing the evaluation field and the preset scene dimension into a first database according to the triplet of the dimension-field-relation value.

Optionally, the field processing unit 32 is specifically configured to:

judging whether the first evaluation field and the second evaluation field are obtained through the same external data interface, wherein the first evaluation field and the second evaluation field are any two fields in the evaluation fields;

if yes, the relation value between the first evaluation field and the second evaluation field is marked as 0, and if not, the relation value between the first evaluation field and the second evaluation field is marked as 1;

judging the number of dimensions hit by the first evaluation field and the second evaluation field together, and adding 1 to the relation value between the first evaluation field and the second evaluation field when one dimension hit by each time;

and storing the relation value between the evaluation fields into a second database according to the triples of the field 1-field 2-relation values.

Optionally, before the field processing unit 32 determines the relationship between the evaluation field in the external data interface and each dimension in the preset scene dimensions, the apparatus for improving the evaluation accuracy further includes:

and the word segmentation unit is used for utilizing the ES word segmentation device to segment the evaluation field.

Alternatively, the map generation unit 33 is specifically configured to:

based on the stored classification results in the first database and the second database, a relational graph of the evaluation field is generated by utilizing the Neo4j graph database.

Optionally, after the evaluation field determining unit 34 obtains a field set having an evaluation influence on the target object greater than a preset influence, the apparatus for improving evaluation accuracy further includes:

and the evaluation unit is used for evaluating the object to be evaluated by using the field set when the object to be evaluated belongs to the target object.

The device for improving the evaluation accuracy provided by the embodiment of the invention can execute the method for improving the evaluation accuracy provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

Fig. 4 is a schematic structural diagram of an apparatus for improving accuracy of evaluation according to an embodiment of the present invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 4, the electronic device 40 includes at least one processor 41, and a memory communicatively connected to the at least one processor 41, such as a Read Only Memory (ROM) 42, a Random Access Memory (RAM) 43, etc., in which the memory stores a computer program executable by the at least one processor, and the processor 41 may perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM) 42 or the computer program loaded from the storage unit 48 into the Random Access Memory (RAM) 43. In the RAM43, various programs and data required for the operation of the electronic device 40 may also be stored. The processor 41, the ROM42 and the RAM43 are connected to each other via a bus 44. An input/output (I/O) interface 45 is also connected to bus 44.

Various components in electronic device 40 are connected to I/O interface 45, including: an input unit 46 such as a keyboard, a mouse, etc.; an output unit 47 such as various types of displays, speakers, and the like; a storage unit 48 such as a magnetic disk, an optical disk, or the like; and a communication unit 49 such as a network card, modem, wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The processor 41 may be various general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 41 performs the various methods and processes described above, such as a method that improves the accuracy of the evaluation.

In some embodiments, the method of improving accuracy of the assessment may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 40 via the ROM42 and/or the communication unit 49. When the computer program is loaded into RAM43 and executed by processor 41, one or more steps of the method of improving accuracy of the assessment described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the method of improving the accuracy of the evaluation in any other suitable way (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

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

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.

Embodiments of the present invention also provide a computer program product comprising computer executable instructions for performing the method of improving accuracy of an assessment provided by any of the embodiments of the present invention when executed by a computer processor.

Of course, the computer program product provided by the embodiments of the present application, whose computer executable instructions are not limited to the method operations described above, may also perform the relevant operations in the method provided by any of the embodiments of the present invention.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.

The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims

1. A method for improving accuracy of an assessment, the method comprising:

2. The method of claim 1, wherein the cleaning the evaluation field comprises:

3. The method for improving evaluation accuracy according to claim 2, wherein classifying the cleaned evaluation field according to a preset scene dimension to obtain a classification result comprises:

judging the relation between the evaluation field in the external data interface and each dimension in the preset scene dimension, and storing the judging result into a first database;

4. The method for improving accuracy of evaluation according to claim 3, wherein determining a relationship between the evaluation field in the external data interface and each of the preset scene dimensions, and storing the determination result in the first database comprises:

judging whether the evaluation field in the external data interface falls into one or more dimensions in the preset scene dimensions;

and storing the evaluation field and the preset scene dimension into the first database according to the triplet of the dimension-field-relation value.

5. The method for improving accuracy of evaluation according to claim 3, wherein determining the relationship between the evaluation fields, and storing the determination result in the second database comprises:

judging whether a first evaluation field and a second evaluation field are obtained through the same external data interface, wherein the first evaluation field and the second evaluation field are any two fields in the evaluation fields;

judging the number of dimensions hit by the first evaluation field and the second evaluation field together, wherein each time a dimension hit by the first evaluation field and the second evaluation field together, the relation value between the first evaluation field and the second evaluation field is increased by 1;

and storing the relation value between the evaluation fields into the second database according to the triplet of the field 1-field 2-relation value.

6. The method of claim 3, wherein prior to determining the relationship between the evaluation field in the external data interface and each of the preset scene dimensions, the method further comprises:

and utilizing an ES word segmentation device to segment the evaluation field.

7. The method of claim 3, wherein generating a relationship graph from the evaluation field using a graph database based on the classification result comprises:

and generating a relation map of the evaluation field by utilizing a Neo4j graphic database based on the stored classification results in the first database and the second database.

8. The method for improving evaluation accuracy according to claim 1, wherein after obtaining a field set having an evaluation influence on the target object greater than a preset influence, the method further comprises:

and when the object to be evaluated belongs to the target object, evaluating the object to be evaluated by utilizing the field set.

9. An apparatus for improving accuracy of an assessment, the apparatus comprising:

10. An apparatus for improving accuracy of evaluation, wherein the apparatus for improving accuracy of evaluation comprises:

one or more processors;

a storage means for storing one or more programs;

when executed by the one or more processors, causes the one or more processors to implement the method of improving accuracy of a rating as set forth in any of claims 1-8.

11. A computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a method of improving accuracy of an assessment according to any of claims 1-8.

12. A computer program product comprising a computer program which, when executed by a processor, implements the method of improving accuracy of an assessment according to any one of claims 1-8.