CN114092057A - Project model construction method and device, terminal equipment and storage medium - Google Patents
Project model construction method and device, terminal equipment and storage medium Download PDFInfo
- Publication number
- CN114092057A CN114092057A CN202111396722.XA CN202111396722A CN114092057A CN 114092057 A CN114092057 A CN 114092057A CN 202111396722 A CN202111396722 A CN 202111396722A CN 114092057 A CN114092057 A CN 114092057A
- Authority
- CN
- China
- Prior art keywords
- model
- label
- combination
- target
- labels
- 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.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 52
- 238000000605 extraction Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000004590 computer program Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000013473 artificial intelligence Methods 0.000 abstract description 7
- 238000007726 management method Methods 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 102000003712 Complement factor B Human genes 0.000 description 7
- 108090000056 Complement factor B Proteins 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 101100545272 Caenorhabditis elegans zif-1 gene Proteins 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
- G06F16/90344—Query processing by using string matching techniques
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application is suitable for the technical field of artificial intelligence, and provides a project model construction method, a project model construction device, a terminal device and a storage medium. The method comprises the following steps: acquiring a data source text associated with a target project model to be constructed; performing structured field extraction processing on the data source text to obtain a plurality of labels; dividing the plurality of labels into more than one label combination according to the categories of the plurality of labels; for each label combination, determining a model factor corresponding to the label combination and the weight of the model factor in the target item model according to the label contained in the label combination; and constructing and obtaining the target project model according to the model factor corresponding to each label combination and the weight of the model factor in the target project model. Compared with the traditional method for manually setting the model factor category and the weight, the method provided by the application can effectively improve the efficiency of constructing the project model.
Description
Technical Field
The application relates to the technical field of artificial intelligence, and provides a project model construction method, a project model construction device, a terminal device and a storage medium.
Background
In the field of financial investment, people usually construct various project models for assisting investment decision-making, and the quality of one project model can directly influence the investment result. At present, the conventional project model construction method is as follows: and (3) manually completing factor planning and sorting of the project model in software such as EXCEL and the like, setting factor categories and weights, and then customizing, developing and realizing. However, the method is complicated to operate, and the project model is constructed with low efficiency.
Disclosure of Invention
In view of this, the present application provides a project model construction method, a project model construction device, a terminal device, and a storage medium, which can improve project model construction efficiency.
In a first aspect, an embodiment of the present application provides a method for constructing a project model, including:
acquiring a data source text associated with a target project model to be constructed;
performing structured field extraction processing on the data source text to obtain a plurality of labels;
dividing the plurality of labels into more than one label combination according to the categories of the plurality of labels;
for each label combination, determining a model factor corresponding to the label combination and the weight of the model factor in the target item model according to the labels contained in the label combination;
and constructing and obtaining the target project model according to the model factor corresponding to each label combination and the weight of the model factor in the target project model.
According to the method and the device, the labels related to the construction of the project model are extracted from the data source text in a structured field extraction mode, then the labels are grouped to obtain each label combination, and the corresponding model factor and the weight of the model factor are obtained based on the labels contained in each label combination, so that the corresponding project model is constructed. Compared with the traditional method for manually setting the model factor category and the weight, the method provided by the application can effectively improve the efficiency of constructing the project model.
In an embodiment of the application, the dividing the plurality of tags into one or more tag combinations according to the categories of the plurality of tags may include:
aiming at any target label in the plurality of labels, extracting a non-numerical character string of the target label;
respectively matching the non-numerical character strings with the characteristic character strings corresponding to the preset label categories to obtain the character string matching degree of each preset label category;
and dividing the target label into label combinations corresponding to label categories with the highest character string matching degree in all the preset label categories.
In an embodiment of the present application, the determining, for each tag combination, a model factor corresponding to the tag combination and a weight of the model factor in the target item model according to the tag included in the tag combination may include:
for any target label combination in the more than one label combination, searching a model factor corresponding to a label included in the target label combination, and determining the searched model factor as the model factor corresponding to the target label combination;
and determining the weight of the model factor corresponding to the target label combination in the target item model according to the number of the labels contained in the target label combination.
Further, the determining, according to the number of tags included in the target tag combination, the weight of the model factor corresponding to the target tag combination in the target item model may include:
counting the total number of the labels contained in the more than one label combination;
calculating the proportion of the number of the labels contained in the target label combination in the total number of the labels;
and determining the proportion as the weight of the model factor corresponding to the target label combination in the target item model.
In an embodiment of the present application, after determining, for each tag combination, a model factor corresponding to the tag combination and a weight of the model factor in the target item model according to the tag included in the tag combination, the method may further include:
scoring each label contained in each label combination according to preset scoring logic to obtain a score of each label;
respectively counting the sum of the scores of all the labels contained in each label combination to obtain the total score of the labels of each label combination;
and adjusting the weight of the model factor corresponding to each label combination in the target item model according to the total label score of each label combination.
In an embodiment of the application, the constructing the target item model according to the model factor corresponding to each label combination and the weight of the model factor in the target item model may include:
acquiring a corresponding general project model according to the category of the target project model, wherein the general project model comprises a preset algorithm, a preset model factor and a preset factor weight;
and replacing the preset model factor by using the model factor corresponding to each label combination, and replacing the preset factor weight by using the weight of the model factor corresponding to each label combination in the target item model to obtain the target item model.
In an embodiment of the present application, after performing a structured field extraction process on the data source text to obtain a plurality of tags, the method may further include:
acquiring index information corresponding to the target project model;
and deleting the labels which are not associated with the index information in the plurality of labels.
In a second aspect, an embodiment of the present application provides a project model building apparatus, including:
the data source text acquisition module is used for acquiring a data source text associated with a target project model to be constructed;
the structured field extraction module is used for executing structured field extraction processing on the data source text to obtain a plurality of labels;
the label combination dividing module is used for dividing the labels into more than one label combination according to the categories of the labels;
a model factor and weight determining module, configured to determine, for each tag combination, a model factor corresponding to the tag combination and a weight of the model factor in the target item model according to a tag included in the tag combination;
and the project model construction module is used for constructing and obtaining the target project model according to the model factor corresponding to each label combination and the weight of the model factor in the target project model.
In a third aspect, an embodiment of the present application provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor, when executing the computer program, implements the method for constructing a project model as set forth in the first aspect of the embodiment of the present application.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored, and the computer program, when executed by a processor, implements the project model building method as set forth in the first aspect of the embodiment of the present application.
In a fifth aspect, the present application provides a computer program product, which when running on a terminal device, causes the terminal device to execute the method for building a project model as set forth in the first aspect of the present application.
The advantageous effects achieved by the second aspect to the fifth aspect described above can be referred to the description of the first aspect described above.
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 embodiments or the prior art descriptions will be briefly described 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 exercise.
FIG. 1 is a flowchart of an embodiment of a method for building a project model provided in an embodiment of the present application;
FIG. 2 is a schematic diagram illustrating an operation principle of a project model construction method according to an embodiment of the present disclosure;
fig. 3 is a block diagram of an embodiment of a project model building apparatus according to an embodiment of the present application;
fig. 4 is a schematic diagram of a terminal device according to an embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail. Furthermore, in the description of the present application and the appended claims, the terms "first," "second," "third," and the like are used for distinguishing between descriptions and not necessarily for describing or implying relative importance.
The quality of a project model in the investment field can directly influence the investment result, so the construction of the project model is very important. How to quickly build, verify, iterate, and publish project models in a system is both a user's expectation and a problem that investment software vendors have to face, who quickly finds and verifies a method first may dominate the overall market. The general project model construction method is to complete factor planning and arrangement of a model in Excel software and other software according to the idea of a user or the model which is actually used, set factor categories and weights, then perform customized development and realization, and perform regression verification in an information system. When the user's idea changes or has a new idea, the iteration is repeated according to the previous method, and obviously, the efficiency of constructing the project model by adopting the method is low.
In view of this, the present application provides a project model construction method, which can improve the project model construction efficiency.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
It should be understood that an execution subject of the project model construction method provided in the embodiment of the present application may be a terminal device or a server, such as a mobile phone, a tablet computer, a wearable device, an in-vehicle device, an Augmented Reality (AR)/Virtual Reality (VR) device, a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, a Personal Digital Assistant (PDA), a large screen television, and the like, and the embodiment of the present application does not set any limit to specific types of the terminal device and the server. The server may be an independent server, or may be a cloud server that provides basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a web service, cloud communication, a middleware service, a domain name service, a security service, a Content Delivery Network (CDN), a big data and artificial intelligence platform, and the like.
Referring to fig. 1, a first embodiment of a method for constructing a project model in an embodiment of the present application includes:
101. acquiring a data source text associated with a target project model to be constructed;
the target project model is a project model to be built, one project model usually comprises a preset algorithm, a plurality of set model factors, the weight of each model factor in the project model and other factors, and by adopting the method provided by the application, the model factors of the project model and the weights of the model factors can be automatically generated, so that the target project model is efficiently built.
First, data source text associated with a target project model is obtained, the data source text being textual material associated with the target project model. Assuming that the user wants to construct a project model of a financial investment class, text material related to the financial investment field, such as the latest analytical articles of the financial investment class, or historical investment index data, etc., can be obtained as the data source text, and it is understood that the data source text can include one or more different text materials.
102. Performing structured field extraction processing on the data source text to obtain a plurality of labels;
after the data source text is obtained, extraction processing of the structured field may be performed on the data source text, so as to obtain a plurality of tags, and the tags may be incorporated into a tag pool for unified management. Specifically, the smallest-grained structured fields (e.g., the industry taxonomy to which the target project model belongs, which may include multiple levels of taxonomy) within the system may be extracted as the respective labels. The extracted labels can be roughly divided into the following two types according to the attribute types of the labels: (1) quantitative labels, such as: various index values (e.g., tax return, pre-amortization profit, internal profitability, etc.), various index grades (e.g., grades 1-5, etc.); (2) qualitative labels, such as: yes, no.
In an implementation manner of the embodiment of the present application, after performing structured field extraction processing on the data source text to obtain a plurality of tags, the method may further include:
(1) acquiring index information corresponding to the target project model;
(2) and deleting the labels which are not associated with the index information in the plurality of labels.
The index information is used for indicating various indexes to be output when the target project model is used, such as return on investment, investment cycle and the like. In actual operation, the association relationship between each item model index and each tag may be preset, after obtaining the plurality of tags, each tag is detected whether to be associated with at least one index of the indexes to be output when the target item model is used, if so, the tag is retained, and if not, the tag is deleted. The labels extracted by adopting the structured fields are many and complicated, a part of labels possibly exist in the labels are irrelevant to all indexes of the target project model, and the part of irrelevant labels are deleted, so that the rationality of constructing the target project model can be improved, and the calculation amount is reduced to a certain extent.
103. Dividing the plurality of labels into more than one label combination according to the categories of the plurality of labels;
after extracting a plurality of labels, dividing label combinations according to the label categories, wherein each label combination comprises at least one label, and each label combination corresponds to one label category. It should be noted that the label category herein does not refer to the attribute category of the label itself, but refers to a preset category into which each label is classified according to a certain rule. For example, if a tag category is preset as a tag combination of financial categories, all tags related to financial categories in the plurality of tags (tag pool) may be classified into the tag combination, and the classified tags may include quantitative tags or qualitative tags. In addition, the same label can be classified into a plurality of different label categories, that is, into a plurality of different label combinations, for example, the labels "yes" and "no", and can be simultaneously classified into a financial label combination and a wind control label combination. In actual operation, the expert user can divide the labels into different label combinations according to the categories and the characteristics of the labels, or can classify the labels by a common user and review the labels by the expert user to confirm that the labels contained in each label combination can embody the characteristics of the label combination.
In an implementation manner of the embodiment of the present application, the automatically classifying the tags may be implemented, and the dividing the tags into more than one tag combination according to the category of the tags may include:
(1) aiming at any target label in the plurality of labels, extracting a non-numerical character string of the target label;
(2) respectively matching the non-numerical character strings with the characteristic character strings corresponding to the preset label categories to obtain the character string matching degree of each preset label category;
(3) and dividing the target label into label combinations corresponding to label categories with the highest character string matching degree in all the preset label categories.
For example, for a certain target label "industry profit margin 30%" in the plurality of labels, firstly, the non-numerical character string "industry profit margin" is extracted, and then, the "industry profit margin" is matched with the feature character string corresponding to each preset label category (such as financial category, wind control category, project management category, and the like), so as to obtain the character string matching degree corresponding to each preset label category. The characteristic character strings corresponding to each label category are preset, for example, a financial category can correspond to a plurality of characteristic character strings such as 'gross profit rate', 'profit margin', 'cost', and the like, after the non-numerical character string 'industry gross profit rate' is matched with all the characteristic character strings corresponding to the financial category, the matching degree of the characteristic character strings 'gross profit rate' and 'industry gross profit rate' is found to be the highest, and then the matching degree of the 'gross profit rate' and the 'industry gross profit rate' can be determined as the character string matching degree corresponding to the financial category; by analogy, the matching degree of the character strings corresponding to all the preset label categories can be obtained. Finally, if the matching degree of the character strings corresponding to the finance category is the highest in all the preset label categories, the target label with the industry gross profit rate of 30 percent can be divided into the label combinations of the finance category. By analogy, the same processing manner as the target label can be performed on each label in the plurality of labels, so that the automatic classification of the plurality of labels is completed. In addition, in order to improve the accuracy of label classification, after the automatic classification is completed, the expert user can also check and correct the classification result.
104. For each label combination, determining a model factor corresponding to the label combination and the weight of the model factor in the target item model according to the label contained in the label combination;
after the plurality of labels are partitioned into label combinations, each label combination may generate a corresponding model factor and a weight of the model factor in the target item model. Each label combination corresponds to a preset label category, that is, the finally generated target item model includes the model factor and the factor weight corresponding to each label category. Assume that there are 3 tag combinations in total for the finance class, the wind control class, and the project management class, where the finance class tag combination includes 3 tags "tag 1, tag 4, and tag 7", the wind control class tag combination includes 2 tags "tag 2 and tag 5", and the project management class tag combination includes 5 tags "tag 3, tag 6, tag 8, tag 9, and tag 10". For example, according to "tag 1, tag 4 and tag 7" included in the tag combination of the financial class, the model factor a corresponding to the financial class and the weight of the model factor a in the target item model can be determined; according to the label 2 and the label 5 contained in the wind control class label combination, the model factor B corresponding to the wind control class and the weight of the model factor B in the target item model can be determined; according to the label 3, the label 6, the label 8, the label 9 and the label 10 included in the item management class label combination, the model factor C corresponding to the item management class and the weight of the model factor C in the target item model can be determined. Specifically, the corresponding relationship between each label and each model factor may be preset, and the weight of each model factor may be set according to a default value, or may be set according to each corresponding label. For the default value, the importance levels of the model factor a, the model factor B, and the model factor C (the importance levels of the respective model factors may be preset by an expert user) may be set on average or may be set, for example, the weight of the model factor a may be set to 50%, the weight of the model factor B may be set to 30%, and the weight of the model factor C may be set to 20%.
In an implementation manner of the embodiment of the present application, for each tag combination, determining, according to a tag included in the tag combination, a model factor corresponding to the tag combination and a weight of the model factor in the target item model may include:
(1) searching a model factor corresponding to a label included in the target label combination aiming at any one target label combination in the more than one label combinations, and determining the searched model factor as the model factor corresponding to the target label combination;
(2) and determining the weight of the model factor corresponding to the target label combination in the target item model according to the number of the labels contained in the target label combination.
The correspondence between each label and the model factor may be preset, for example: the model factor "industry rating" may correspond to the label "industry scale XX, industry gross profit ratio XX, industry growth ratio XX" (XX represents a certain numerical value). For the above example, the model factor a corresponding to the financial class and the weight of the model factor a in the target item model may be determined according to "tag 1, tag 4 and tag 7" included in the financial class tag combination, specifically, the corresponding weight may be determined according to the number of tags (3) included in the financial class tag combination, and generally, the larger the number of tags, the larger the weight may be set.
Further, the determining, according to the number of tags included in the target tag combination, the weight of the model factor corresponding to the target tag combination in the target item model may include:
(1) counting the total number of the labels contained in the more than one label combination;
(2) calculating the proportion of the number of the labels contained in the target label combination in the total number of the labels;
(3) and determining the proportion as the weight of the model factor corresponding to the target label combination in the target item model.
For example, in the above example, the 3 tag combinations of the finance class, the wind control class and the project management class include 10 tags in total, wherein the tag combination of the finance class includes 3 tags, and the proportion is 30%, so the weight of the model factor a corresponding to the tag combination of the finance class in the target project model can be set to 30%. By analogy, the weight of the model factor B corresponding to the wind control type tag combination in the target project model is 20%, and the weight of the model factor C corresponding to the project management type tag combination in the target project model is 50%.
In an implementation manner of the embodiment of the present application, after determining, for each tag combination, a model factor corresponding to the tag combination and a weight of the model factor in the target item model according to a tag included in the tag combination, the method may further include:
(1) scoring each label contained in each label combination according to preset scoring logic to obtain a score of each label;
(2) respectively counting the sum of the scores of all the labels contained in each label combination to obtain the total score of the labels of each label combination;
(3) and adjusting the weight of the model factor corresponding to each label combination in the target item model according to the total label score of each label combination.
For step (1), corresponding scoring logics may be respectively set for the labels of different attribute categories, for example, for a quantitative label "industry gross profit rate 35%", a corresponding relationship between a numerical range and a score may be set, for example, more than 50% corresponds to 100 points, 40% -50% corresponds to 90 points, 30% -40% corresponds to 80 points …, and then the score of the label "industry gross profit rate 35%" may be determined to be 80; for a quantitative label "risk level 2", a corresponding relationship between a numerical level and a score may be set, for example, a level 0 corresponds to 100 points, a level 1 corresponds to 80 points, and a level 2 corresponds to 60 points …, so that the score of the label "risk level 2" may be determined to be 60 points; for qualitative labels "yes" and "no," yes "may be set for 100 points, no" for 0 points, and so on.
For step (2), the total score of the label of each label combination is calculated respectively, specifically, the sum of the scores of all the labels contained in the label combination is calculated. For example, for the financial class label combination in the above example, which includes label 1, label 4 and label 7, if the score of label 1 is 100, the score of label 4 is 0, and the score of label 7 is 50, the total score of the labels of the financial class label combination can be calculated to be 150.
For step (3), according to the total label score of the label combination, the weight of the corresponding model factor in the target item model can be adjusted. The specific adjustment criteria may be: if the total label score of a certain label combination is higher than a set threshold value, the weight of the corresponding model factor in the target item model can be increased according to a certain proportion; if the total label score of a certain label combination is lower than a set threshold, the weight of the corresponding model factor in the target item model can be reduced according to a certain proportion. The adjusted weight may be set according to a ratio of the total label score of each label combination to the sum of the total label scores of all label combinations. For example, in the above example, if the total label score of the finance type label combination is 150, the total label score of the wind control type label combination is 300, and the total label score of the project management type label combination is 150, the sum of the total label scores of all the label combinations may be determined to be 600, where the total label score of the finance type label combination is 25%, so the weight of the model factor a corresponding to the finance type label combination in the target project model may be reduced to 25%, and so on. And adjusting the corresponding model factor weight according to the score value of each label, so that the reasonability and the effectiveness of the constructed project model can be improved to a certain extent.
105. And constructing and obtaining the target project model according to the model factor corresponding to each label combination and the weight of the model factor in the target project model.
After determining each model factor and the weight of each model factor contained in the target project model, the target project model can be constructed. For example, in the above example, a target item model may be constructed that includes 3 model factors (model factor a, model factor B, and model factor C), and the weight of each model factor is a corresponding value (model factor a is 30%, model factor B is 20%, and model factor C is 50%).
In an implementation manner of the embodiment of the present application, the constructing the target item model according to the model factor corresponding to each label combination and the weight of the model factor in the target item model may include:
(1) acquiring a corresponding general project model according to the category of the target project model, wherein the general project model comprises a preset algorithm, a preset model factor and a preset factor weight;
(2) and replacing the preset model factor by using the model factor corresponding to each label combination, and replacing the preset factor weight by using the weight of the model factor corresponding to each label combination in the target item model to obtain the target item model.
In actual operation, a plurality of different types of general project models can be set, and each general project model is provided with a corresponding algorithm, a default model factor and a factor weight. According to the type of the target project model to be constructed, for example, the financial investment class, a general project model of the financial investment class may be obtained, then the model factor of the general project model is replaced by using the model factor determined in step 104, and the model factor weight of the general project model is replaced by using the model factor weight determined in step 104, so as to obtain the target project model. Through the arrangement, the corresponding project model can be constructed only by executing the replacement of the model factors and the factor weights, the workload of model construction can be effectively reduced, and the efficiency of project model construction is improved.
According to the method and the device, the labels related to the construction of the project model are extracted from the data source text in a structured field extraction mode, then the labels are grouped to obtain each label combination, and the corresponding model factor and the weight of the model factor are obtained based on the labels contained in each label combination, so that the corresponding project model is constructed. Compared with the traditional method for manually setting the factor category and the weight of the model, the method provided by the application can effectively improve the efficiency of constructing the project model.
Fig. 2 is a schematic diagram illustrating an operation principle of the project model construction method proposed in the present application. In fig. 2, first, various tags are extracted from the data source text by means of structured field extraction, and the tags are stored in a tag pool. Then, the labels in the label pool are classified to obtain label combinations, wherein each label combination comprises more than one label. Next, for each tag combination, a corresponding model factor and a weight for the model factor may be generated, respectively. And finally, constructing a corresponding project model based on the obtained model factors and the weights of the model factors.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
Fig. 3 is a block diagram showing a construction apparatus of a project model according to an embodiment of the present application, which corresponds to the method of constructing a project model according to the above embodiment, and only shows a part related to the embodiment of the present application for convenience of description.
Referring to fig. 3, the apparatus includes:
a data source text acquisition module 301, configured to acquire a data source text associated with a target project model to be constructed;
a structured field extraction module 302, configured to perform structured field extraction processing on the data source text to obtain a plurality of tags;
a label combination dividing module 303, configured to divide the plurality of labels into more than one label combination according to the category of the plurality of labels;
a model factor and weight determining module 304, configured to determine, for each tag combination, a model factor corresponding to the tag combination and a weight of the model factor in the target item model according to a tag included in the tag combination;
a project model building module 305, configured to build the target project model according to the model factor corresponding to each tag combination and the weight of the model factor in the target project model.
In one embodiment of the present application, the tag combination partitioning module may include:
a non-numeric character string extraction unit, configured to extract, for any target tag among the multiple tags, a non-numeric character string of the target tag;
the character string matching unit is used for matching the non-numerical character strings with the characteristic character strings corresponding to the preset label categories respectively to obtain the character string matching degree of each preset label category;
and the label dividing unit is used for dividing the target label into a label combination corresponding to the label category with the highest character string matching degree in the preset label categories.
In one embodiment of the present application, the model factor and weight determination module may include:
a model factor searching unit, configured to search, for any target label combination in the more than one label combinations, a model factor corresponding to a label included in the target label combination, and determine the searched model factor as a model factor corresponding to the target label combination;
and the factor weight determining unit is used for determining the weight of the model factor corresponding to the target label combination in the target item model according to the number of labels contained in the target label combination.
Further, the factor weight determination unit may include:
the tag number counting subunit is used for counting the total number of tags contained in the more than one tag combination;
a label number ratio calculating unit, configured to calculate a ratio of the number of labels included in the target label combination to the total number of labels;
and the weight determining subunit is used for determining the proportion as the weight of the model factor corresponding to the target label combination in the target item model.
In an embodiment of the application, the project model building apparatus may further include:
the label scoring module is used for scoring each label contained in each label combination according to preset scoring logic to obtain the score of each label;
the label score counting module is used for respectively counting the sum of scores of all labels contained in each label combination to obtain the total score of the labels of each label combination;
and the factor weight adjusting module is used for adjusting the weight of the model factor corresponding to each label combination in the target item model according to the total label score of each label combination.
In one embodiment of the present application, the project model building module may include:
a general item model obtaining unit, configured to obtain a corresponding general item model according to a category of the target item model, where the general item model includes a preset algorithm, a preset model factor, and a preset factor weight;
and the model factor and weight replacing unit is used for replacing the preset model factor by using the model factor corresponding to each label combination and replacing the preset factor weight by using the weight of the model factor corresponding to each label combination in the target item model to obtain the target item model.
In an embodiment of the present application, the project model building apparatus may further include:
the index information acquisition module is used for acquiring index information corresponding to the target project model;
and the label deleting module is used for deleting the labels which are not associated with the index information in the plurality of labels.
Embodiments of the present application further provide a computer-readable storage medium, where computer-readable instructions are stored, and when executed by a processor, the computer-readable instructions implement any one of the project model building methods shown in fig. 1.
The embodiment of the present application further provides a computer program product, when the computer program product runs on a server, the server is caused to execute a building method for implementing any one of the project models as represented in fig. 1.
Fig. 4 is a schematic diagram of a terminal device according to an embodiment of the present application. As shown in fig. 4, the terminal device 4 of this embodiment includes: a processor 40, a memory 41, and computer readable instructions 42 stored in the memory 41 and executable on the processor 40. The processor 40, when executing the computer readable instructions 42, implements the steps in the above-described embodiments of the method of building a project model, such as the steps 101 to 105 shown in fig. 1. Alternatively, the processor 40, when executing the computer readable instructions 42, implements the functions of the modules/units in the above device embodiments, such as the functions of the modules 301 to 305 shown in fig. 3.
Illustratively, the computer readable instructions 42 may be partitioned into one or more modules/units that are stored in the memory 41 and executed by the processor 40 to accomplish the present application. The one or more modules/units may be a series of computer-readable instruction segments capable of performing specific functions, which are used for describing the execution process of the computer-readable instructions 42 in the terminal device 4.
The terminal device 4 may be a computing device such as a smart phone, a notebook, a palm computer, and a cloud terminal device. The terminal device 4 may include, but is not limited to, a processor 40 and a memory 41. It will be understood by those skilled in the art that fig. 4 is only an example of the terminal device 4, and does not constitute a limitation to the terminal device 4, and may include more or less components than those shown, or combine some components, or different components, for example, the terminal device 4 may further include an input-output device, a network access device, a bus, etc.
The Processor 40 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an AppLication Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 41 may be an internal storage unit of the terminal device 4, such as a hard disk or a memory of the terminal device 4. The memory 41 may also be an external storage device of the terminal device 4, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, provided on the terminal device 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the terminal device 4. The memory 41 is used for storing the computer readable instructions and other programs and data required by the terminal device. The memory 41 may also be used to temporarily store data that has been output or is to be output.
It should be noted that, for the information interaction, execution process, and other contents between the above devices/units, the specific functions and technical effects thereof based on the same concept as those of the method embodiment of the present application can be specifically referred to the method embodiment portion, and are not described herein again.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium and can implement the steps of the embodiments of the methods described above when the computer program is executed by a processor. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a photographing apparatus/terminal apparatus, a recording medium, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium. Such as a usb-disk, a removable hard disk, a magnetic or optical disk, etc.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (10)
1. A project model construction method is characterized by comprising the following steps:
acquiring a data source text associated with a target project model to be constructed;
performing structured field extraction processing on the data source text to obtain a plurality of labels;
dividing the plurality of labels into more than one label combination according to the categories of the plurality of labels;
for each label combination, determining a model factor corresponding to the label combination and the weight of the model factor in the target item model according to the label contained in the label combination;
and constructing and obtaining the target project model according to the model factor corresponding to each label combination and the weight of the model factor in the target project model.
2. The method of claim 1, wherein the dividing the plurality of labels into more than one label combination according to the categories of the plurality of labels comprises:
aiming at any target label in the plurality of labels, extracting a non-numerical character string of the target label;
respectively matching the non-numerical character strings with the characteristic character strings corresponding to the preset label categories to obtain the character string matching degree of each preset label category;
and dividing the target label into label combinations corresponding to label categories with the highest character string matching degree in all the preset label categories.
3. The method of claim 1, wherein for each tag combination, determining a model factor corresponding to the tag combination and a weight of the model factor in the target item model according to the tags included in the tag combination comprises:
searching a model factor corresponding to a label included in the target label combination aiming at any one target label combination in the more than one label combinations, and determining the searched model factor as the model factor corresponding to the target label combination;
and determining the weight of the model factor corresponding to the target label combination in the target item model according to the number of the labels contained in the target label combination.
4. The method of claim 3, wherein the determining the weight of the model factor corresponding to the target tag combination in the target item model according to the number of tags contained in the target tag combination comprises:
counting the total number of the labels contained in the more than one label combination;
calculating the proportion of the number of the labels contained in the target label combination in the total number of the labels;
and determining the proportion as the weight of the model factor corresponding to the target label combination in the target item model.
5. The method of claim 1, wherein after determining, for each of the tag combinations, a model factor corresponding to the tag combination and a weight of the model factor in the target item model according to the tags included in the tag combination, further comprising:
scoring each label contained in each label combination according to preset scoring logic to obtain a score of each label;
respectively counting the sum of the scores of all the labels contained in each label combination to obtain the total score of the labels of each label combination;
and adjusting the weight of the model factor corresponding to each label combination in the target item model according to the total label score of each label combination.
6. The method of claim 1, wherein the constructing the target item model according to the model factor corresponding to each label combination and the weight of the model factor in the target item model comprises:
acquiring a corresponding general project model according to the category of the target project model, wherein the general project model comprises a preset algorithm, a preset model factor and a preset factor weight;
and replacing the preset model factor by using the model factor corresponding to each label combination, and replacing the preset factor weight by using the weight of the model factor corresponding to each label combination in the target item model to obtain the target item model.
7. The method of any of claims 1 to 6, wherein after performing a structured field extraction process on the data source text resulting in a plurality of tags, further comprising:
acquiring index information corresponding to the target project model;
and deleting the labels which are not associated with the index information in the plurality of labels.
8. An apparatus for constructing a project model, comprising:
the data source text acquisition module is used for acquiring a data source text associated with a target project model to be constructed;
the structured field extraction module is used for executing structured field extraction processing on the data source text to obtain a plurality of labels;
the label combination dividing module is used for dividing the labels into more than one label combination according to the categories of the labels;
a model factor and weight determining module, configured to determine, for each tag combination, a model factor corresponding to the tag combination and a weight of the model factor in the target item model according to a tag included in the tag combination;
and the project model construction module is used for constructing and obtaining the target project model according to the model factor corresponding to each label combination and the weight of the model factor in the target project model.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the project model construction method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, implements the method of constructing a project model according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111396722.XA CN114092057A (en) | 2021-11-23 | 2021-11-23 | Project model construction method and device, terminal equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111396722.XA CN114092057A (en) | 2021-11-23 | 2021-11-23 | Project model construction method and device, terminal equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114092057A true CN114092057A (en) | 2022-02-25 |
Family
ID=80303436
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111396722.XA Pending CN114092057A (en) | 2021-11-23 | 2021-11-23 | Project model construction method and device, terminal equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114092057A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116562649A (en) * | 2023-05-12 | 2023-08-08 | 实道时代(北京)科技有限公司 | Investment decision management method, system and storage medium based on artificial intelligence |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111932267A (en) * | 2020-07-27 | 2020-11-13 | 中国工商银行股份有限公司 | Enterprise financial service risk prediction method and device |
| CN112000748A (en) * | 2020-07-14 | 2020-11-27 | 北京神州泰岳智能数据技术有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN112579771A (en) * | 2020-12-08 | 2021-03-30 | 腾讯科技（深圳）有限公司 | Content title detection method and device |
| CN113554184A (en) * | 2021-07-22 | 2021-10-26 | 上海晓途网络科技有限公司 | Model training method and device, electronic equipment and storage medium |
-
2021
- 2021-11-23 CN CN202111396722.XA patent/CN114092057A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112000748A (en) * | 2020-07-14 | 2020-11-27 | 北京神州泰岳智能数据技术有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN111932267A (en) * | 2020-07-27 | 2020-11-13 | 中国工商银行股份有限公司 | Enterprise financial service risk prediction method and device |
| CN112579771A (en) * | 2020-12-08 | 2021-03-30 | 腾讯科技（深圳）有限公司 | Content title detection method and device |
| CN113554184A (en) * | 2021-07-22 | 2021-10-26 | 上海晓途网络科技有限公司 | Model training method and device, electronic equipment and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116562649A (en) * | 2023-05-12 | 2023-08-08 | 实道时代(北京)科技有限公司 | Investment decision management method, system and storage medium based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108846520B (en) | Loan overdue prediction method, loan overdue prediction device and computer-readable storage medium | |
| CN111177569A (en) | Recommendation processing method, device and equipment based on artificial intelligence | |
| CN109409677A (en) | Enterprise Credit Risk Evaluation method, apparatus, equipment and storage medium | |
| CN110569322A (en) | Address information analysis method, device and system and data acquisition method | |
| CN112328909B (en) | Information recommendation method and device, computer equipment and medium | |
| CN110765117A (en) | Fraud identification method and device, electronic equipment and computer-readable storage medium | |
| CN115391669B (en) | Intelligent recommendation method and device and electronic equipment | |
| CN111324738B (en) | Method and system for determining text label | |
| CN110310012B (en) | Data analysis method, device, equipment and computer readable storage medium | |
| CN110852785B (en) | User grading method, device and computer readable storage medium | |
| CN113592605B (en) | Product recommendation method, device, equipment and storage medium based on similar products | |
| CN108572988A (en) | A kind of house property assessment data creation method and device | |
| CN110782349A (en) | Model training method and system | |
| CN111738762A (en) | Method, device, equipment and storage medium for determining recovery price of poor assets | |
| CN112328657A (en) | Feature derivation method, feature derivation device, computer equipment and medium | |
| CN113807728A (en) | Performance assessment method, device, equipment and storage medium based on neural network | |
| CN115204971A (en) | Product recommendation method and device, electronic equipment and computer-readable storage medium | |
| CN114860941A (en) | Industry data management method and system based on data brain | |
| CN114092057A (en) | Project model construction method and device, terminal equipment and storage medium | |
| CN113344125A (en) | Long text matching identification method and device, electronic equipment and storage medium | |
| CN112163098A (en) | Knowledge graph creating method and device, storage medium and server | |
| CN116843483A (en) | Vehicle insurance claim settlement method, device, computer equipment and storage medium | |
| CN114708073B (en) | Intelligent detection method and device for surrounding mark and serial mark, electronic equipment and storage medium | |
| CN116701752A (en) | News recommendation method and device based on artificial intelligence, electronic equipment and medium | |
| CN117011577A (en) | Image classification method, apparatus, computer device and storage medium |
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 |