CN110858214B - Recommendation model training and further auditing program recommendation method, device and equipment - Google Patents
Recommendation model training and further auditing program recommendation method, device and equipment Download PDFInfo
- Publication number
- CN110858214B CN110858214B CN201810961351.7A CN201810961351A CN110858214B CN 110858214 B CN110858214 B CN 110858214B CN 201810961351 A CN201810961351 A CN 201810961351A CN 110858214 B CN110858214 B CN 110858214B
- Authority
- CN
- China
- Prior art keywords
- audit
- program
- preset data
- recommendation model
- recommendation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
Abstract
The application discloses a recommendation model training and further auditing program recommendation method, device and equipment, comprising: acquiring audit history data in a preset format; extracting text contents respectively corresponding to preset data items from audit historical data; the preset data items comprise audit analysis conclusions and further audit procedures; abstracting the text content corresponding to each extracted preset data item by using the first keyword list to obtain an abstract description text corresponding to each preset data item; and training the pre-established recommendation model by using the abstract description texts corresponding to the preset data items respectively to obtain the trained recommendation model, wherein the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program. In addition, recommendations for further audit procedures are made using the trained recommendation model. The method and the device can accurately and efficiently complete the further determination of the auditing program, and improve the overall quality of the auditing service.
Description
Technical Field
The application relates to the field of auditing, in particular to a recommendation model training and further auditing program recommendation method, device and equipment.
Background
In the auditing practice, after the risk assessment program and the auditing analysis are implemented by the auditor, an auditing analysis conclusion is obtained, and a risk point exists in the auditing analysis conclusion, and in order to minimize the risk point, the auditing analysis conclusion needs to be verified by adopting a further auditing program so as to ensure the accuracy of the auditing analysis conclusion obtained by the auditor.
Assuming that the audit analysis conclusion obtained by the auditor is cost-questionable, in order to ensure the accuracy of the audit analysis conclusion, further audit procedures need to be performed on each risk point which may cause cost-questionable, for example, further accounting is performed on the cost, and if the audit analysis conclusion is proved to be correct after the further audit procedures are performed, the persuasiveness of the audit analysis conclusion is ensured.
At present, the selection of the further auditing program based on the auditing analysis conclusion is manually completed by the auditing staff, and due to the limitation of factors of professional ability, working experience and the like of the auditing staff, the selection of the further auditing program is inevitable to cause various problems, such as insufficient or irrelevant correlation between the further auditing program and an auditing risk point, omission of important substantive programs, unreasonable combination of the further auditing program and the like, and the overall quality of the auditing service is seriously influenced.
Disclosure of Invention
In order to solve the problems, the recommendation model training and further audit program recommendation method, device and equipment are provided, the recommendation model is trained, the recommendation model is used for more accurately and efficiently determining the further audit program, and the overall quality of the audit service is improved. The specific technical scheme is as follows:
in a first aspect, the present application provides a method for further auditing program recommendations, the method comprising:
acquiring audit history data in a preset format;
extracting text contents respectively corresponding to preset data items from the audit historical data; the text content corresponding to each preset data item has a corresponding relation, and the preset data items comprise audit analysis conclusions and further audit programs;
abstracting the text content corresponding to each extracted preset data item by using the first keyword list to obtain an abstract description text corresponding to each preset data item; the first keyword list stores keywords corresponding to each preset data item; and training the pre-established recommendation model by using the abstract description texts corresponding to the preset data items respectively to obtain the trained recommendation model, wherein the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program.
Optionally, the preset data item further includes audit verification items, where the audit verification items are used to describe an audit service scenario; correspondingly, the recommendation model is specifically used for describing the relationship among the audit verification items, the audit analysis conclusion and the further audit program.
Optionally, the extracting, from the audit history data, text contents corresponding to preset data items respectively includes:
performing word segmentation processing on the audit historical data to obtain the audit historical data subjected to word segmentation processing;
searching in the audit history data after word segmentation processing by using a second keyword list to obtain text contents corresponding to each preset data item; and the second keyword list stores retrieval keywords corresponding to all preset data items respectively.
In a second aspect, an embodiment of the present application further provides a further auditing program recommendation method, where the method includes:
and taking any audit analysis conclusion and/or audit verification matters corresponding to the audit analysis conclusion as input parameters of the trained recommendation model in any recommendation model training method of the first aspect, and outputting a further audit program corresponding to the audit analysis conclusion and/or the audit verification matters after the recommendation model is processed.
Optionally, the further auditing procedure output by the recommendation model includes a further auditing procedure combination with a sequential relationship.
In a third aspect, the present application provides a recommendation model training apparatus, including:
the acquisition module is used for acquiring audit historical data in a preset format;
the extraction module is used for extracting text contents corresponding to preset data items from the audit historical data; the text contents corresponding to each preset data item have a corresponding relationship, and the preset data items comprise audit analysis conclusions and further audit programs;
the abstraction module is used for abstracting the text content corresponding to each extracted preset data item by using the first keyword list to obtain an abstraction description text corresponding to each preset data item; the first keyword list stores keywords corresponding to each preset data item;
and the training module is used for training the pre-established recommendation model by using the abstract description texts corresponding to the preset data items respectively to obtain the trained recommendation model, wherein the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program.
Optionally, the preset data item further includes audit verification items, where the audit verification items are used to describe an audit service scenario; correspondingly, the recommendation model is specifically used for describing the relationship among the audit verification items, the audit analysis conclusion and the further audit program.
Optionally, the extracting module includes:
the word segmentation processing submodule is used for carrying out word segmentation processing on the audit historical data to obtain the audit historical data after word segmentation processing;
the retrieval submodule is used for retrieving the audit history data subjected to word segmentation by using a second keyword list to obtain text contents corresponding to the preset data items respectively; and the second keyword list stores retrieval keywords corresponding to all preset data items respectively.
In a fourth aspect, the present application further provides a further auditing program recommendation apparatus, including:
and the recommending module is used for taking any audit analysis conclusion and/or audit verification matters corresponding to the audit analysis conclusion as input parameters of the trained recommending model in any recommending model training device in the third aspect, and outputting further auditing programs corresponding to the audit analysis conclusion and the audit verification matters after the processing of the recommending model.
Optionally, the further auditing procedure output by the recommendation model includes a further auditing procedure combination with a sequential relationship.
In a fifth aspect, the present application provides a storage medium comprising a stored program, wherein the program performs the method of any one of the first or second aspects.
In a sixth aspect, the present application provides a processor for executing a program, wherein the program executes to perform the method of any one of the first or second aspects.
According to the recommendation model training method and the further audit program recommendation method, mass audit historical data are analyzed in a machine learning mode, a recommendation model is established, recommendation of further audit programs is achieved through the recommendation model trained through a large number of samples, and compared with the prior art, recommendation of further audit programs can be completed more accurately and efficiently, and the overall quality of audit services is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.
Fig. 1 is a flowchart of a recommendation model training method according to an embodiment of the present application;
FIG. 2 is a flow chart of a further auditing program recommendation method provided by an embodiment of the present application;
fig. 3 is a schematic structural diagram of a recommended model training apparatus according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a further auditing program recommending apparatus according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Aiming at the audit analysis conclusion made by the auditor, a further audit program is adopted for verification, so that the accuracy of the audit analysis conclusion provided by the auditor can be ensured, and the selection of the further audit program becomes a key point for ensuring the accuracy of the audit analysis conclusion, and the accuracy of the selection of the further audit program cannot be ensured in the conventional mode of manually selecting the further audit program based on the auditor, so that the application provides a recommendation model training method, a further audit program recommendation method, a device and equipment, which are used for analyzing mass audit historical data by using a machine learning mode, establishing a recommendation model, training the recommendation model by using a large number of samples, realizing the recommendation of the further audit program by using the trained recommendation model, and being capable of more accurately and efficiently completing the recommendation of the further audit program compared with the prior art, and the overall quality of the audit service is improved.
An embodiment of a recommended model training method provided by the present application is specifically described below, and with reference to fig. 1, a flowchart of the recommended model training method provided by the embodiment of the present application is specifically described, where the method specifically includes:
s101: and acquiring audit historical data in a preset format.
The audit history data has a wide range, and may generally include an enterprise internal audit work draft, a business annual report audit draft, a special audit work draft, and the like.
The form of the audit history data is various, so the format of the audit history data is also different, and the audit history data can generally comprise the audit history data in the formats of Excel documents, Word documents, e-mails, databases and the like. In order to facilitate subsequent utilization of the audit history data, the obtained audit history data in various formats need to be sorted to obtain the audit history data in the preset format. The preset format may include an Excel document, a Word document, and the like, and specifically, the format is arranged according to the content in the audit history data, the text content in the audit history data may be arranged into a Word document, and the table content may be arranged into an Excel document, and the like.
S102: extracting text contents respectively corresponding to preset data items from the audit historical data; the text contents corresponding to the preset data items respectively have corresponding relations, and the preset data items comprise audit analysis conclusions and further audit procedures.
After the audit history data in the preset format is obtained, the audit history data is further sorted, and specifically, text contents corresponding to the preset data items are extracted from the audit history data.
The preset data items may include an audit analysis conclusion and a further audit program, where the audit analysis conclusion is a conclusion obtained after an auditor performs a risk assessment program and an audit analysis in audit history data, for example, the audit analysis conclusion may be "settlement problem", "return on investment rate 60%", "illegal issue individual difference rate fee", and the like, where the audit analysis conclusion has risk points, for example, the "settlement problem" may be caused by a raw material inventory error, or may be caused by a management fee settlement problem, and the like, so that various risk points of the audit analysis conclusion have corresponding further audit programs, for example, the raw material may be rechecked for the raw material inventory error, and the management fee may be rechecked for the management fee settlement problem. The above are only a few simple examples, and in fact, various audit analysis conclusions can exist in the audit practice, and various risk points need to be adopted for further auditing by various further auditing programs. The further auditing procedure refers to further verification that is taken against the risk points in order to ensure the accuracy of the audit analysis conclusions drawn by the auditors, wherein the selection of the further auditing procedure needs to be sufficiently related to the risk points.
In practical application, firstly, word segmentation processing is performed on the audit history data in the preset format acquired in the step S101, so as to obtain the audit history data after word segmentation processing. Specifically, the word segmentation processing for Chinese generally adopts a double natural language processing algorithm, a Chinese word segmentation algorithm, a Hadamard LTP, a Chinese academy NLPIR and other word segmentation processing algorithms; the English word segmentation processing generally adopts TF-IDF (term frequency-inverse text frequency index) or BM25(Best Match 25) and other algorithms. Secondly, respectively setting a second keyword list for the preset data items in advance, wherein the second keyword list stores the retrieval keywords of each preset data item, and text contents corresponding to the preset data items can be retrieved from the audit historical data through the retrieval keywords. Specifically, the second keyword list may be stored in a structure in which each preset data item corresponds to its own search keyword, for example, in a two-column table in which the preset data item and the search keyword are stored respectively. And searching in the audit history data after word segmentation processing by using the search keywords of the preset data items to obtain the text content corresponding to each data item. In an optional implementation manner, a certain section of text, in the audit history data after word segmentation, of which the matching degree with the keywords of any data item reaches a preset value is taken as the text content corresponding to the data item.
It is worth noting that the text content corresponding to each preset data item can be deduplicated to finally obtain structured data. Through the implementation mode, at least the text contents corresponding to the audit analysis conclusion and the further audit program can be extracted from the audit historical data, and the text contents corresponding to the audit analysis conclusion and the further audit program are extracted and stored in a corresponding relation, for example, the text content corresponding to the audit analysis conclusion is that settlement is problematic due to raw material inventory error, the text content corresponding to the further audit program is that raw material is rechecked, and specifically, the settlement is problematic due to raw material inventory error and the raw material is rechecked again are stored in a corresponding relation.
In an optional implementation manner, the preset data item may further include audit verification items, where the audit verification items are used for describing an audit service scenario. For one audit analysis conclusion, the audit analysis conclusion can belong to a plurality of audit service scenes, the same audit analysis conclusion belonging to different audit service scenes corresponds to different further audit programs, and the audit analysis conclusion under a specific audit service scene can correspond to more accurate further audit programs. In addition, other data items may also be set according to requirements in the embodiment of the present application, which is not described herein again.
S103: abstracting the text content corresponding to each extracted preset data item by using the first keyword list to obtain an abstract description text corresponding to each preset data item; the first keyword list comprises keywords corresponding to all preset data items.
Because the sources of the text contents corresponding to the preset data items are different, and the description languages of different auditors or other workers are generally different, the text contents corresponding to the preset data items are abstracted, so that the text contents expressing the same meaning use a uniform expression language, and more standardized text description is obtained.
In practical application, a first keyword list is preset for each preset data item, wherein keywords respectively corresponding to each preset data item are stored in the first keyword list, and the text contents of the preset data items can be abstracted by comparing the keywords with the text contents corresponding to the preset data items to obtain an abstracted description text, so that the text contents with different description languages but the same expression meaning are used in a uniform expression language, and the text description is more standardized. Specifically, the first keyword list may be stored in a structure in which each preset data item corresponds to its own keyword, for example, in a two-column table in which the preset data item and the keyword are stored respectively. Specifically, the first keyword list may be the same as the second keyword list, that is, the keywords of each preset data item are the same as the search keywords; otherwise, it may be different. In an optional implementation manner, the keywords corresponding to each preset data item may form a keyword combination (including one or more keywords), each keyword combination corresponds to an abstract description text, and if the matching degree between the text content corresponding to a certain preset data item and any keyword combination in the preset data item reaches a preset value, the abstract description text corresponding to the keyword combination replaces the original text content, so as to implement normalized text description. For example, there may be three expressions for "implement a letter to a supplier": the method comprises the steps of sending a letter to a supplier, judging whether the amount of an accounts payable of the letter is accurate, sending an inquiry letter to the supplier with a plan account balance exceeding 50 ten thousand, checking whether the balance of related subjects is really existed and accurate through the inquiry letter, and uniformly expressing the letter as an 'implement letter to the supplier' serving as an abstract description text through the abstract processing to realize standardized text description.
In practical application, after the text content corresponding to each preset data item is abstracted, the obtained abstract description text can be subjected to deduplication processing, so that the data is more refined.
S104: and training the pre-established recommendation model by using the abstract description texts corresponding to the preset data items respectively to obtain the trained recommendation model, wherein the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program.
In the embodiment of the application, a recommendation model for recommending further auditing programs is established in advance, and specifically, the establishment of the recommendation model can be completed by establishing a specific mathematical function or model. The established recommendation model can be used for describing the relationship between the audit analysis conclusion and the further audit program, for example, the probability or the frequency of selecting a certain further audit program for a certain audit analysis conclusion, and the like.
After the recommendation model is established, the recommendation model needs to be trained. Because the abstract description texts respectively corresponding to the audit analysis conclusion and the further audit program can be obtained from a large amount of audit historical data by using the above-mentioned S101-S103, the abstract description texts having the corresponding relationship can be used as the recommendation model training samples in the embodiment of the present application, so as to implement training of the recommendation model, and the trained recommendation model can be used for accurately describing the relationship between the audit analysis conclusion and the further audit program.
In an optional implementation manner, if the preset data item includes an audit analysis conclusion and a further audit program, and also includes audit verification items, the abstract description texts respectively corresponding to the three acquired from a large amount of audit historical data are used as training samples of the recommendation model, so that training of the recommendation model is realized, and the trained recommendation model can be used for accurately describing the relationships between the audit verification items, the audit analysis conclusion and the further audit program.
In practical application, the recommendation model trained by the method can be used for recommending further audit procedures in audit work, so that the further audit procedures can be determined more accurately and efficiently, and the overall quality of audit services is improved.
Specifically, an embodiment of the present application further provides a recommendation method for a further audit program, and with reference to fig. 2, on the basis of the above steps in fig. 1, the recommendation method for a further audit program provided in the embodiment of the present application further includes step S105, which is specifically as follows: s105: and taking any audit analysis conclusion and/or audit verification matters corresponding to the audit analysis conclusion as input parameters of the trained recommendation model in the recommendation model training method, and outputting a further audit program corresponding to the audit analysis conclusion and/or the audit verification matters after the recommendation model is processed.
In the embodiment of the application, after the training of the recommendation model is completed, the trained recommendation model can be used for recommending the further audit program, and the recommendation model is trained by a large number of training samples, so that the purpose of more efficiency and accuracy can be achieved by using the recommendation model to recommend the further audit program, and the quality of the audit service is improved on the whole.
In practical application, if the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program, any audit analysis conclusion can be used as an input parameter of the recommendation model, and after the recommendation model is processed, the further audit program is output, wherein the further audit program is the further audit program adopted by the input audit analysis conclusion.
If the recommendation model is used for describing the audit verification items and the relationship between the audit analysis conclusion and the further audit program, the audit verification items corresponding to any audit analysis conclusion and the audit analysis conclusion can be used as input parameters of the recommendation model, and after the audit verification items are processed by the recommendation model, the further audit program is output, and the further audit program is the further audit program adopted by the input audit analysis conclusion under the audit verification items.
In the recommendation model training method and the further audit program recommendation method provided by the embodiment of the application, mass audit historical data are analyzed in a machine learning mode, a recommendation model is established, a large number of samples are used for training the recommendation model, the trained recommendation model is used for realizing the recommendation of a further audit program, compared with the prior art, the recommendation of the further audit program can be completed more accurately and efficiently, and the overall quality of audit service is improved.
In an alternative embodiment, the output of the recommendation model may also be a further auditing program combination with a sequential relationship, where the further auditing program combination may exist in the form of a further auditing program list, where the further auditing program list refers to that the recommendation model may be used by a user through recommending a plurality of further auditing programs for the user. The further auditing programs included in the further auditing program list can be sorted according to the adopted times, and the further auditing programs with the adopted times are preferentially provided for the user to adopt. In addition, the further audit program combination may also refer to a group of further audit programs having an execution sequence, and each further audit program in the combination is regarded as one program in a specific execution process, and needs to be executed once according to the sequence.
In addition, the method and the device can also receive the adjustment of the sequence of the further auditing program or the further auditing program combination recommended by the user to the recommendation model, and obtain the adjusted further auditing program or the further auditing program combination. The scoring module of the system preferably scores the further auditing program or the further auditing program combination recommended by the recommendation model (namely, the first score), and specific scoring rules include but are not limited to the following modes: the recommendation model multiplies 100 the ratio of the number of occurrences of a particular further audit program recommended for a particular audit analysis conclusion (or audit analysis conclusion and audit verification event) to the total number of occurrences of all further audit programs or further audit program combinations. When the recommendation result of the recommendation model is not satisfied by the user, the ranking of the further auditing program or the further auditing program combination can be adjusted and scored (i.e. the second score), and the specific scoring rule includes but is not limited to the following modes: the reciprocal of the ratio of the total number of occurrences of a further audit program or a combination of further audit programs selected by a user for a certain audit analysis conclusion (or audit analysis conclusion and audit verification event) to the number of adjustments made by the user to the further audit program or the combination of further audit programs under the recommendation model is multiplied by 100. The scoring of the user may affect the scoring of the recommendation model, in practical applications, the scores are collected respectively, and a weight coefficient (a first weight coefficient and a second weight coefficient, the sum of which is 1) is manually set for each score according to specific situations, to calculate a final recommendation ranking score, and a further auditing program combination provided by the recommendation model for the user and having a sequential relationship may be several further auditing programs with higher recommendation ranking scores.
By the aid of the method, when the recommendation model is used for recommending the further audit programs, the recommendation model can recommend the first N further audit programs with higher scores for the user, and the user can select the required further audit programs.
Corresponding to the above method embodiment, the present application also provides a further audit program recommendation apparatus, and referring to fig. 3, a schematic structural diagram of the further audit program recommendation apparatus provided in the embodiment of the present application is provided, where the apparatus includes:
an obtaining module 201, configured to obtain audit history data in a preset format;
an extracting module 202, configured to extract text contents corresponding to preset data items from the audit history data; the text contents corresponding to each preset data item have a corresponding relationship, and the preset data items comprise audit analysis conclusions and further audit programs;
the abstraction module 203 is configured to perform abstraction processing on the extracted text content corresponding to each preset data item by using the first keyword list, so as to obtain an abstraction description text corresponding to each preset data item; the first keyword list comprises keywords corresponding to all preset data items respectively;
the training module 204 is configured to train a pre-established recommendation model by using the abstraction description texts corresponding to the preset data items, so as to obtain a trained recommendation model, where the recommendation model is used to describe a relationship between the audit analysis conclusion and the further audit program.
The preset data item also comprises audit verification items, and the audit verification items are used for describing an audit service scene; correspondingly, the recommendation model is specifically used for describing the relationship among the audit verification items, the audit analysis conclusion and the further audit program.
The extraction module comprises:
the word segmentation processing submodule is used for carrying out word segmentation processing on the audit historical data to obtain the audit historical data after word segmentation processing;
the retrieval submodule is used for retrieving the audit history data subjected to word segmentation by using a second keyword list to obtain text contents corresponding to each preset data item; the second keyword list comprises retrieval keywords corresponding to all preset data items respectively.
In addition, on the basis of each module of the recommendation model training device, the embodiment of the present application further provides a further audit program recommendation device, and on the basis of each module of the recommendation model training device, the embodiment of the present application further includes a recommendation module 205, and referring to fig. 4, a schematic structural diagram of the further audit program recommendation device provided by the embodiment of the present application is provided.
The recommending module 205 is specifically configured to:
and taking any audit analysis conclusion and/or audit verification matters corresponding to the audit analysis conclusion as input parameters of the trained recommendation model in any recommendation model training device, and outputting a further audit program corresponding to the audit analysis conclusion and the audit verification matters after the recommendation model is processed.
The further auditing procedure of the recommendation model output comprises a further auditing procedure combination with a sequence relation.
In the recommendation model training device and the further audit program recommendation device provided by the embodiment of the application, mass audit historical data are analyzed in a machine learning mode, a recommendation model is established, the recommendation model is trained through a large amount of sample training, the recommendation of the further audit program is realized through the trained recommendation model, compared with the prior art, the recommendation of the further audit program can be completed more accurately and efficiently, and the overall quality of audit service is improved.
Correspondingly, the recommendation model training device and the further auditing program recommendation device respectively comprise a processor and a memory, the acquisition module, the extraction module, the abstraction module, the training module, the recommendation module and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. One or more kernels can be set, further audit program determination can be completed more accurately and efficiently by adjusting kernel parameters, and the overall quality of audit service is improved.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
An embodiment of the present invention provides a storage medium on which a program is stored, the program implementing the recommendation model training method or the further auditing program recommendation method when executed by a processor.
The embodiment of the invention provides a processor, which is used for running a program, wherein the program executes the recommendation model training method or the further auditing program recommendation method when running.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps:
acquiring audit history data in a preset format;
extracting text contents respectively corresponding to preset data items from the audit historical data; the text content corresponding to each preset data item has a corresponding relation, and the preset data items comprise audit analysis conclusions and further audit programs;
abstracting text contents corresponding to the preset data items by using first keyword lists respectively set for the preset data items to obtain abstract description texts corresponding to the preset data items;
training a pre-established recommendation model by using abstract description texts corresponding to all preset data items respectively to obtain a trained recommendation model, wherein the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program;
and recommending a further auditing program by using the trained recommendation model.
Optionally, the preset data item further includes audit verification items, where the audit verification items are used to describe an audit service scenario; correspondingly, the recommendation model is specifically used for describing the relationship among the audit verification items, the audit analysis conclusion and the further audit program.
Optionally, the extracting, from the audit history data, text contents corresponding to preset data items respectively includes:
performing word segmentation processing on the audit historical data to obtain the audit historical data subjected to word segmentation processing;
and searching the audit history data after word segmentation processing by using second keyword lists respectively set for preset data items to obtain text contents respectively corresponding to the preset data items.
The embodiment of the invention also provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can be operated on the processor, wherein the processor executes the program and realizes the following steps:
taking any audit analysis conclusion and/or audit verification matters corresponding to the audit analysis conclusion as input parameters of the trained recommendation model in any recommendation model training method, and outputting a further audit program corresponding to the audit analysis conclusion and/or the audit verification matters after the recommendation model is processed.
Optionally, the further auditing procedure output by the recommendation model includes a further auditing procedure combination with a sequential relationship.
The device herein may be a server, a PC, a PAD, a mobile phone, etc.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device:
acquiring audit history data in a preset format;
extracting text contents respectively corresponding to preset data items from the audit historical data; the text content corresponding to each preset data item has a corresponding relation, and the preset data items comprise audit analysis conclusions and further audit programs;
abstracting text contents corresponding to the preset data items by using first keyword lists respectively set for the preset data items to obtain abstract description texts corresponding to the preset data items;
training a pre-established recommendation model by using abstract description texts corresponding to all preset data items respectively to obtain a trained recommendation model, wherein the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program;
and recommending a further auditing program by using the trained recommendation model.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device:
and taking any audit analysis conclusion and/or audit verification matters corresponding to the audit analysis conclusion as input parameters of the trained recommendation model in any recommendation model training method, and outputting a further audit program corresponding to the audit analysis conclusion and/or the audit verification matters after the recommendation model is processed.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both permanent and non-permanent, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A recommendation model training method, the method comprising:
acquiring audit history data in a preset format;
extracting text contents respectively corresponding to preset data items from the audit historical data; the text contents corresponding to each preset data item have a corresponding relationship, and the preset data items comprise audit analysis conclusions and further audit programs;
abstracting the text content corresponding to each extracted preset data item by using the first keyword list to obtain an abstract description text corresponding to each preset data item; the first keyword list stores keywords corresponding to each preset data item;
and training the pre-established recommendation model by using the abstract description texts corresponding to the preset data items respectively to obtain the trained recommendation model, wherein the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program.
2. The method of claim 1, wherein the preset data items further comprise audit verification items, and the audit verification items are used for describing audit service scenarios; correspondingly, the recommendation model is specifically used for describing the relationship among the audit verification items, the audit analysis conclusion and the further audit program.
3. The method according to claim 1, wherein the extracting, from the audit history data, text contents corresponding to preset data items respectively comprises:
performing word segmentation processing on the audit historical data to obtain the audit historical data subjected to word segmentation processing;
searching in the audit history data after word segmentation processing by using a second keyword list to obtain text contents corresponding to each preset data item; and the second keyword list stores retrieval keywords corresponding to all preset data items respectively.
4. A method of further auditing program recommendations, the method comprising:
taking any audit analysis conclusion and/or audit verification matters corresponding to the audit analysis conclusion as input parameters of the trained recommendation model in the recommendation model training method of any claim 1 to 3, and outputting a further audit program corresponding to the audit analysis conclusion and/or the audit verification matters after the recommendation model is processed.
5. A further auditing program recommendation method according to claim 4, wherein the further auditing program of recommendation model output comprises a further auditing program combination with a sequential relationship.
6. A recommendation model training apparatus, the apparatus comprising:
the acquisition module is used for acquiring audit historical data in a preset format;
the extraction module is used for extracting text contents corresponding to preset data items from the audit historical data; the text contents corresponding to each preset data item have corresponding relations, and the preset data items comprise audit analysis conclusions and further audit programs;
the abstraction module is used for abstracting the text content corresponding to each extracted preset data item by using the first keyword list to obtain an abstraction description text corresponding to each preset data item; the first keyword list stores keywords corresponding to each preset data item;
and the training module is used for training the pre-established recommendation model by using the abstract description texts corresponding to the preset data items respectively to obtain the trained recommendation model, wherein the recommendation model is used for describing the relationship between the audit analysis conclusion and the further audit program.
7. The apparatus of claim 6, wherein the preset data item further comprises an audit verification item, and the audit verification item is used for describing an audit service scenario; correspondingly, the recommendation model is specifically used for describing the relationship among the audit verification items, the audit analysis conclusion and the further audit program.
8. A further audit program recommendation apparatus, the apparatus comprising:
a recommending module, configured to take any audit analysis conclusion and/or audit verification items corresponding to the audit analysis conclusion as input parameters of the trained recommending model in the recommending model training apparatus of claim 6 or 7, and output a further auditing procedure corresponding to the audit analysis conclusion and the audit verification items after processing by the recommending model.
9. A storage medium, characterized in that the storage medium comprises a stored program, wherein the program performs the method of any one of claims 1-5.
10. A processor, characterized in that the processor is configured to run a program, wherein the program when running performs the method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810961351.7A CN110858214B (en) | 2018-08-22 | 2018-08-22 | Recommendation model training and further auditing program recommendation method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810961351.7A CN110858214B (en) | 2018-08-22 | 2018-08-22 | Recommendation model training and further auditing program recommendation method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110858214A CN110858214A (en) | 2020-03-03 |
| CN110858214B true CN110858214B (en) | 2022-07-05 |
Family
ID=69636084
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810961351.7A Active CN110858214B (en) | 2018-08-22 | 2018-08-22 | Recommendation model training and further auditing program recommendation method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110858214B (en) |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6134664A (en) * | 1998-07-06 | 2000-10-17 | Prc Inc. | Method and system for reducing the volume of audit data and normalizing the audit data received from heterogeneous sources |
| US7353228B2 (en) * | 2000-12-07 | 2008-04-01 | General Electric Capital Corporation | Method and product for calculating a net operating income audit and for enabling substantially identical audit practices among a plurality of audit firms |
| US7715930B2 (en) * | 2006-09-27 | 2010-05-11 | Rockwell Automation Technologies, Inc. | Aggregating audit information with field conditions |
| US8762180B2 (en) * | 2009-08-25 | 2014-06-24 | Accenture Global Services Limited | Claims analytics engine |
| WO2016145089A1 (en) * | 2015-03-09 | 2016-09-15 | Skytree, Inc. | System and method for using machine learning to generate a model from audited data |
| US11188648B2 (en) * | 2015-08-12 | 2021-11-30 | Micro Focus Llc | Training a security scan classifier to learn an issue preference of a human auditor |
| CN105160038B (en) * | 2015-10-10 | 2017-04-19 | 广东卓维网络有限公司 | Data analysis method and system based on audit database |
| CN106815125A (en) * | 2015-12-02 | 2017-06-09 | 阿里巴巴集团控股有限公司 | A kind of log audit method and platform |
| CN107577737A (en) * | 2017-08-25 | 2018-01-12 | 北京百度网讯科技有限公司 | Method and apparatus for pushed information |
| CN107832429A (en) * | 2017-11-14 | 2018-03-23 | 广州供电局有限公司 | audit data processing method and system |
-
2018
- 2018-08-22 CN CN201810961351.7A patent/CN110858214B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110858214A (en) | 2020-03-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11887010B2 (en) | Data classification for data lake catalog | |
| US11762926B2 (en) | Recommending web API's and associated endpoints | |
| CN110334356A (en) | Article matter method for determination of amount, article screening technique and corresponding device | |
| US11755663B2 (en) | Search activity prediction | |
| US9996529B2 (en) | Method and system for generating dynamic themes for social data | |
| CN112149387A (en) | Visualization method and device for financial data, computer equipment and storage medium | |
| CN107368489B (en) | Information data processing method and device | |
| Zhang et al. | Cslabel: An approach for labelling mobile app reviews | |
| CN110765101A (en) | Label generation method and device, computer readable storage medium and server | |
| CN110689211A (en) | Method and device for evaluating website service capability | |
| CN110188291B (en) | Document processing based on proxy log | |
| CN110544035A (en) | internal control detection method, system and computer readable storage medium | |
| US20210192125A1 (en) | Methods and systems for facilitating summarization of a document | |
| CN115982429B (en) | Knowledge management method and system based on flow control | |
| CN109542890B (en) | Data modification method, device, computer equipment and storage medium | |
| CN110858214B (en) | Recommendation model training and further auditing program recommendation method, device and equipment | |
| CN108520012A (en) | Mobile Internet user comment method for digging based on machine learning | |
| Battanta et al. | Regtech: Case studies of cooperation with banks in italy | |
| Sumner et al. | RipetaScore: Measuring the quality, transparency, and trustworthiness of a scientific work | |
| Ma et al. | API prober–a tool for analyzing web API features and clustering web APIs | |
| Lin et al. | Realtime event summarization from tweets with inconsistency detection | |
| CN111782601A (en) | Electronic file processing method and device, electronic equipment and machine readable medium | |
| CN116225770B (en) | Patch matching method, device, equipment and storage medium | |
| CN113077272B (en) | Communication business marketing scheme optimization method and device | |
| CN113704103B (en) | Test case recommendation method and device, medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |