CN116883175A - Investment and financing activity decision generation method and device - Google Patents
Investment and financing activity decision generation method and device Download PDFInfo
- Publication number
- CN116883175A CN116883175A CN202310843148.0A CN202310843148A CN116883175A CN 116883175 A CN116883175 A CN 116883175A CN 202310843148 A CN202310843148 A CN 202310843148A CN 116883175 A CN116883175 A CN 116883175A
- Authority
- CN
- China
- Prior art keywords
- decision
- financing
- investment
- information
- activity
- 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.)
- Granted
Links
- 230000000694 effects Effects 0.000 title claims abstract description 151
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000003066 decision tree Methods 0.000 claims abstract description 113
- 238000005516 engineering process Methods 0.000 claims abstract description 31
- 238000012549 training Methods 0.000 claims abstract description 20
- 238000012800 visualization Methods 0.000 claims abstract description 15
- 238000013138 pruning Methods 0.000 claims description 24
- 238000004422 calculation algorithm Methods 0.000 claims description 22
- 230000000007 visual effect Effects 0.000 claims description 18
- 230000008901 benefit Effects 0.000 claims description 7
- 230000033228 biological regulation Effects 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 25
- 238000005457 optimization Methods 0.000 abstract description 4
- 238000004458 analytical method Methods 0.000 description 13
- 238000007726 management method Methods 0.000 description 13
- 238000012795 verification Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 6
- 238000007794 visualization technique Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000003062 neural network model Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Game Theory and Decision Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Human Resources & Organizations (AREA)
- Operations Research (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method and a device for generating investment and financing activity decisions. The method comprises the following steps: acquiring record information of a plurality of historical financing activities, wherein the record information comprises a plurality of corresponding characteristic information and decision conclusions of each historical financing activity; training to obtain a financing activity decision model, and constructing a corresponding decision tree; acquiring a plurality of corresponding characteristic information of the to-be-decided investment and financing activity, and inputting the characteristic information into an investment and financing activity decision model to obtain a decision conclusion of the to-be-decided investment and financing activity; marking decision branches matched with decision conclusions of the to-be-decided financing activity in the decision tree, and displaying the decision conclusions of the to-be-decided financing activity and the marked decision tree through a visualization technology. The application can understand the thinking process of making decision by the investment and financing activity decision model, not only improves the transparency and the interpretability of the artificial intelligence decision technology, but also facilitates the adjustment and the optimization of the decision by a user.
Description
Technical Field
The application relates to the technical field of finance, in particular to a method and a device for generating investment and financing activity decisions.
Background
Artificial intelligence decision making techniques refer to processes and methods for making decisions using artificial intelligence techniques. The method mainly utilizes artificial intelligence technologies such as machine learning, deep learning, natural language processing, image recognition and the like to analyze and process a large amount of data so as to obtain a decision result. Compared with the traditional decision method, the artificial intelligence decision technology has higher accuracy and efficiency, and can help enterprises and organizations to better cope with complex business and market environments.
Artificial intelligence decision-making techniques may be applied in various fields such as finance, medical, education, transportation, etc. In the financial field, artificial intelligence decision-making technology can be used in stock investment, risk assessment, credit rating and other aspects; in the medical field, the artificial intelligence decision-making technology can be used for disease diagnosis, drug research and development, medical resource allocation and other aspects; in the education field, the artificial intelligence decision-making technology can be used for student assessment, course design, teaching management and other aspects; in the traffic field, the artificial intelligence decision-making technology can be used for traffic flow control, road condition monitoring, intelligent navigation and other aspects.
There are also challenges and problems in the use of artificial intelligence decision making techniques. The method mainly comprises the following steps:
1. transparency and interpretability problems: the black box problem of artificial intelligence decision technology has been the focus of attention, and how to interpret and understand decision results is an important challenge of artificial intelligence decision technology. This is because many artificial intelligence algorithms employ complex mathematical models and algorithms that are often difficult to understand and interpret.
Specifically, when the intelligent analysis is used for managing and deciding the investment and financing, because the intelligent analysis model is equivalent to a black box, people can only directly obtain the result output by the intelligent analysis model, but it is difficult to understand how the algorithm in the intelligent analysis model obtains the decision result, so that the transparency and the interpretability of the artificial intelligent decision technology are poor, and uncertainty and untrustworthy feel are brought to people.
2. Data quality problem: artificial intelligence decision-making techniques require a large amount of data support, but the quality of the data has a crucial impact on the decision outcome. If there is an error, missing or deviation in the data, this will lead to inaccuracy in the decision result.
3. Algorithm and model selection problem: different algorithms and models are suitable for different scenes and problems, and the selection of the proper algorithm and model is important for the accuracy and reliability of the decision result.
4. Privacy and security issues: artificial intelligence decision-making techniques require a large amount of data support, and these data often involve sensitive information such as personal privacy and business confidentiality, and how to guarantee the security and privacy of the data is another important challenge of artificial intelligence decision-making techniques.
Disclosure of Invention
Based on the above, in order to solve the technical problems in the prior art, when the intelligent analysis is used to manage and decide the investment and financing, only the result output by the intelligent analysis model can be directly obtained, and it is difficult to understand how the algorithm in the intelligent analysis model obtains the decision result, so that the transparency and the interpretability of the artificial intelligent decision technology are poor.
In order to achieve the above object, the present application provides the following technical solutions:
in a first aspect, a method for generating a financing activity decision includes:
s1, acquiring record information of a plurality of historical financing activities, wherein the record information comprises a plurality of corresponding characteristic information and decision conclusions of each historical financing activity;
s2, training to obtain a financing activity decision model according to the record information of the historical financing activities, and constructing a corresponding decision tree;
s3, obtaining a plurality of corresponding characteristic information of the to-be-decided investment and financing activity, and inputting the plurality of corresponding characteristic information of the to-be-decided investment and financing activity into the investment and financing activity decision model to obtain a decision conclusion of the to-be-decided investment and financing activity;
s4, marking a decision branch matched with the decision conclusion of the to-be-decided financing activity in the decision tree, and displaying the decision conclusion of the to-be-decided financing activity and the decision tree with the mark through a visualization technology.
Optionally, the plurality of characteristic information includes a plurality of market factor information, a plurality of technical factor information, a plurality of financial factor information, a plurality of policy factor information, and a plurality of management factor information.
Further optionally, the plurality of market factor information includes market size, market demand, and market competition, the plurality of technical factor information includes technical difficulty, technical maturity, and technical advantage, the plurality of financial factor information includes cost, expected return on investment of investment projects, the plurality of policy factor information includes government policies and laws and regulations, and the plurality of management factor information includes enterprise management level, team quality, and enterprise culture.
Optionally, step S1 further includes:
judging whether the historical financing activities lack of characteristic information exist in the recorded information of the historical financing activities;
when the historical investment and financing activities with missing feature information are judged to exist, judging whether the number of the historical investment and financing activities with missing feature information exceeds a preset number threshold;
if the number of the historical investment and financing activities with the missing characteristic information is not more than a preset number threshold, deleting the recorded information of the historical investment and financing activities with the missing information;
if the number of the historical investment and financing activities of the missing feature information exceeds the preset number threshold, the value of the missing feature information is given to the common value or the average value of the corresponding feature.
Optionally, the constructing the corresponding decision tree includes:
selecting the characteristic information with the maximum information gain ratio from the recorded information of the plurality of historical financing activities as a root node of a decision tree by using a C4.5 algorithm;
and selecting optimal partition attributes layer by layer through the information gain ratio to construct the decision tree layer by layer.
Optionally, step S2 further includes:
and pruning the decision tree after the decision tree is constructed.
Optionally, in step S4, the decision tree with the label is presented by means of an echartis visual drawing tool.
Further optionally, the presenting the decision tree with the label by the echorts visual drawing tool includes:
defining a JSON object containing data of said decision tree, said JSON object containing all nodes of said decision tree;
creating an empty canvas by using the Tree chart type in the Echarts visual drawing tool, and customizing the appearance of the canvas by setting the attribute of the canvas;
adding the data of the decision tree into the empty canvas by utilizing the JSON object;
defining the style of the node in the canvas by setting the attribute of the node;
the style of an edge in the canvas is defined by setting the properties of the edge.
Optionally, the method further comprises:
respectively acquiring a plurality of corresponding characteristic information of a plurality of to-be-decided investment and financing activities, respectively inputting the corresponding plurality of characteristic information of each to-be-decided investment and financing activity into the investment and financing activity decision model, and obtaining a decision conclusion of each to-be-decided investment and financing activity;
judging the to-be-decided financing campaign which can be used as the optimal choice according to the decision conclusion of each to-be-decided financing campaign;
marking all decision branches in the decision tree, which are matched with the decision conclusion of each to-be-decided financing activity, respectively, and displaying the to-be-decided financing activity which can be used as the optimal selection and the decision tree with the mark through a visualization technology.
In a second aspect, a investment and financing activity decision-making apparatus includes:
the system comprises a record information acquisition module, a decision making module and a decision making module, wherein the record information acquisition module is used for acquiring record information of a plurality of historical financing activities, and the record information comprises a plurality of corresponding characteristic information and decision making conclusions of each historical financing activity;
the model training and decision tree construction module is used for training to obtain a financing activity decision model according to the record information of the historical financing activities and constructing a corresponding decision tree;
the decision generation module is used for acquiring a plurality of corresponding characteristic information of the to-be-decided investment and financing activity, inputting the corresponding plurality of characteristic information of the to-be-decided investment and financing activity into the investment and financing activity decision model, and obtaining a decision conclusion of the to-be-decided investment and financing activity;
the visual display module is used for marking a decision branch matched with the decision conclusion of the to-be-decided financing activity in the decision tree, and displaying the decision conclusion of the to-be-decided financing activity and the decision tree with the mark through a visual technology.
The application has at least the following beneficial effects:
by the method for generating the investment and financing activity decision, the decision tree marked with the decision branch related to the decision conclusion can be obtained while a decision activity to be invested is input into a trained investment and financing activity decision model to obtain the corresponding decision conclusion; after the enterprise obtains the decision conclusion and the decision tree through the visualization technology, the decision conclusion and the decision tree are combined, and the path of the decision object to be invested in the decision tree can be clear, so that the thinking process of the investment and financing activity decision model for making the decision conclusion is understood, how the decision conclusion is obtained by an algorithm in the investment and financing activity decision model is understood, the transparency and the interpretability of the artificial intelligence decision technology are improved, and the adjustment and the optimization of the decision by a user can be facilitated.
Drawings
FIG. 1 is a flow chart of a method for generating a financing activity decision according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a decision tree generation process under a C4.5 algorithm in accordance with one embodiment of the present application;
FIG. 3 is a schematic diagram of a decision tree constructed in accordance with one embodiment of the present application;
FIG. 4 is a schematic flow chart of a method for generating a financing decision according to an embodiment of the present application;
FIG. 5 is a block diagram of a module architecture of a investment and financing activity decision-making apparatus according to an embodiment of the present application;
fig. 6 is an internal structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
In one embodiment, as shown in fig. 1, a method for generating a financing activity decision is provided, comprising the steps of:
s1, acquiring record information of a plurality of historical financing activities, wherein the record information comprises a plurality of corresponding characteristic information and decision conclusions of each historical financing activity.
Wherein the plurality of characteristic information includes a plurality of market factor information, a plurality of technical factor information, a plurality of financial factor information, a plurality of policy factor information, and a plurality of management factor information. Specifically, the plurality of market factor information includes market size, market demand and market competition, the plurality of technical factor information includes technical difficulty, technical maturity and technical advantage, the plurality of financial factor information includes cost, expected return on investment, return on investment of investment projects, the plurality of policy factor information includes government policies and laws and regulations, and the plurality of management factor information includes enterprise management level, team quality and enterprise culture.
In other words, the feature information of all records of the financing decision object is firstly obtained and used as a training sample set of a neural network model and a decision tree, and the training sample set specifically comprises market factor information such as market scale, market demand, market competition and the like, technical factor information such as technical difficulty, technical maturity, technical advantage and the like, financial factor information such as cost, expected income, return on investment and the like of investment projects, policy factor information such as government policies, laws and regulations and the like, and management factor information such as enterprise management level, team quality, enterprise culture and the like.
Further, step S1 further includes:
judging whether historical financing activities with missing characteristic information exist in the recorded information of the plurality of historical financing activities;
when the historical investment and financing activities with missing feature information are judged to exist, judging whether the number of the historical investment and financing activities with missing feature information exceeds a preset number threshold;
if the number of the historical investment and financing activities with the missing characteristic information is not more than a preset number threshold, deleting the recorded information of the historical investment and financing activities with the missing information;
if the number of the historical investment and financing activities of the missing feature information exceeds the preset number threshold, the value of the missing feature information is given to the common value or the average value of the corresponding feature.
Through this step, the algorithm provided by the embodiment of the application can process incomplete data sets.
In other words, in actual operation, a sample set lacking certain attribute values (i.e., feature information values) may be obtained. If the number of samples with missing attribute values in the sample set is small, incomplete samples can be deleted directly. If there are a large number of samples in the dataset that lack attribute values, the samples cannot simply be deleted, since the large number of deleted samples lose a large amount of useful information for the model and decision tree, and the result obtained is not accurate enough. In this case, the correct way to handle missing attribute values is to assign a common value, or attribute mean, to the feature.
S2, training to obtain a financing activity decision model according to the record information of a plurality of historical financing activities, and constructing a corresponding decision tree.
For those skilled in the art, training the decision neural network model for the investment and financing activity belongs to a conventional technical means in the prior art, and is not described herein. The decision neural network model of the financing activity obtained through training can take the characteristic information of the financing activity as input to directly output corresponding decision conclusions, for example, conclusions which are worth focusing, not worth focusing, general focusing and the like can be output.
Further, constructing the corresponding decision tree includes:
selecting characteristic information with the maximum information gain ratio from the recorded information of a plurality of historical financing activities as a root node of a decision tree by using a C4.5 algorithm;
and selecting optimal partition attributes layer by layer through the information gain ratio to construct the decision tree layer by layer.
In particular, decision trees are a commonly used machine learning algorithm that can be used for classification and regression problems. The decision tree builds a tree structure by recursively dividing the data set, each node represents an attribute or feature, each branch represents a value of the attribute or feature, and the leaf nodes represent classification or regression results. In a classification problem, the decision tree may separate the dataset into different categories, while in a regression problem it may predict a continuous value. The decision tree algorithm has the advantages of easy understanding, small calculated amount, capability of processing missing values and the like, and is widely applied to practical application.
The application of the C4.5 algorithm can be expressed specifically as:
after all recorded attributes (namely characteristic information) of the financing decision object are acquired, firstly, the information gain ratio of each attribute is calculated, and then the attribute with the largest information gain ratio is selected as the root node. The information gain ratio is the ratio of the information gain to the attribute inherent value. The information gain refers to the degree to which the information entropy of a sample set decreases after a certain attribute is selected as a division criterion, and the attribute eigenvalue refers to the amount of information contained in one attribute itself. The calculation of the information gain ratio can avoid excessive preference of the attribute with more values, and can also avoid excessive preference of the attribute with less values.
For each attribute, the information gain and the attribute eigenvalue are calculated, the information gain ratio of each attribute is calculated, and the attribute with the largest information gain ratio is selected as the dividing attribute. That is, the attribute with the largest information gain ratio is selected as the dividing attribute of the current non-leaf node, the value of the attribute is taken as a branch, and the next layer of the decision tree is continuously constructed. The above process is repeated for each child node until all leaf nodes are of the same class or cannot be subdivided.
In other words, after the system calculates, the market factor related attribute with the maximum information gain ratio is selected as the root node. After the standard is divided, the information entropy of the sample set is greatly reduced, and after the attribute with the maximum information gain ratio is selected as the root node, the algorithm takes the value of the attribute as a branch to continue to construct the next layer of the decision tree. By analogy, after comprehensively analyzing market factors, technical factors, financial factors, policy factors and management factors, the market prospect of investment, the technical prospect of investment, the economic benefit of investment, the policy environment of investment, the management level of investment and potential risks are defined. Dividing the data set according to the factor information, and dividing the data set of the current node into different subsets according to the selected optimal dividing attribute, wherein each subset corresponds to one branch. A subtree is recursively generated, one for each subset, until the data of all leaf nodes belong to the same class or reach a predetermined maximum depth.
Further, step S2 further includes:
after the decision tree is constructed, pruning is carried out on the decision tree.
When constructing a decision tree, the problem of over fitting easily occurs, so pruning is needed to improve the generalization capability of the decision tree.
As an alternative pruning approach, the recorded information data sets of the plurality of historical financing activities may be divided into a training set and a validation set when constructing the corresponding decision tree, where the training set is used to construct the decision tree and the validation set is used to prune. Pruning is performed from bottom to top starting from the leaf nodes of the decision tree. For each leaf node, the accuracy of pruning the leaf node on the verification set and the accuracy of not pruning the leaf node on the verification set are calculated. If the accuracy after pruning is not inferior to the accuracy without pruning, pruning the leaf node. Otherwise, pruning is not performed.
In other words, after the decision tree is built, pruning is performed on the decision tree. When the decision tree is used for classification, the data set is divided into a training set and a verification set, wherein the training set is used for constructing a decision tree model, and the verification set is used for testing the generalization capability of the model, namely the expressive capability of the model on new data. The accuracy of the verification set of the decision tree can be used as an important index for measuring the generalization capability of the model, and reflects the classification accuracy of the model on new data. Pruning is carried out on the decision tree, when pruning is carried out, each non-leaf node is considered from bottom to top from the leaf node of the decision tree, and if the node is changed into the leaf node, so that the accuracy of the verification set can be improved.
As an alternative pruning method, pessimistic pruning method (Pessimistic ErrorPruning) in pruning can be adopted, and the core idea of the method is to judge whether subtrees can be pruned according to the misjudgment rate before and after pruning. A subtree (with multiple leaf nodes) is classified into one leaf node, so that the false positive rate on the training set must rise, but the false positive rate for a new sample does not necessarily rise. The node can be pruned assuming that the accuracy of a branch node on the test sample does not vary much before and after the node is pruned. While pessimistic pruning has limitations, it performs well in practical applications. In addition, the method does not need to separate a training set and a verification set, and is beneficial to training a data set with smaller scale. And pessimistic pruning is more efficient and faster than other pruning methods. Because each subtree in the tree needs to be asked only once during pruning, in the worst case, too much computation time is not required.
The decision tree generation process under the above-mentioned C4.5 algorithm can also be seen in fig. 2, and a decision tree obtained finally after pruning can be seen in fig. 3.
S3, obtaining a plurality of corresponding characteristic information of the to-be-decided investment and financing activity, and inputting the corresponding plurality of characteristic information of the to-be-decided investment and financing activity into an investment and financing activity decision model to obtain a decision conclusion of the to-be-decided investment and financing activity.
Feature information of the to-be-decided financing activities expected to obtain decision conclusion is input into a trained financing activity decision model, so that a conclusion similar to a conclusion which is worth focusing, not worth focusing or general focusing can be intuitively obtained.
And S4, marking decision branches matched with decision conclusions of the to-be-decided financing activity in the decision tree, and displaying the decision conclusions of the to-be-decided financing activity and the marked decision tree through a visualization technology.
Where decision conclusions are presented, it can be briefly discussed why this conclusion was made. Or, the decision conclusion can be interpreted according to the decision branches matched with the decision conclusion, so that enterprises can better understand the decision conclusion.
Further, in step S4, the decision tree with the label is displayed by the echartis visual drawing tool.
Specifically, exposing the marked decision tree through the Echarts visual drawing tool comprises:
defining a JSON object containing data of the decision tree, wherein the JSON object contains all nodes of the decision tree;
creating an empty canvas by using the Tree chart type in the Echarts visual drawing tool, and customizing the appearance of the canvas by setting the attribute of the canvas;
adding the data of the decision tree into an empty canvas by using the JSON object;
defining the style of the node in the canvas by setting the attribute of the node;
the style of an edge in the canvas is defined by setting the properties of the edge.
Specifically, echartits is a JavaScript-based visualization library with which to visualize the decision tree generated by the presentation:
first, a JSON object containing decision tree data needs to be defined. The JSON object contains a root node and all child nodes, each node having a unique ID and a tag.
Next, an empty canvas needs to be created using the Tree chart type of Echarts. The appearance of the canvas may be defined by setting the size, title, color, etc. properties of the canvas. Decision tree data is then required to be added to the canvas. The setOption method of Echarts may be used to set the data of the graph, including information of the ID of the node, the tag, the parent node ID, etc.
Next, a style of node needs to be defined. The appearance of the node can be defined by setting the shape, color, font and other attributes of the node. The style of a node may be set using the itemStyle and label properties of Echarts.
Finally, the style of the edge needs to be defined. The appearance of the edge can be defined by setting the color, width, line type, etc. of the edge. The style of an edge may be set using the linetype attribute of Echarts.
Thus, the intelligent investment and financing decision making by using the visualization and the decision tree is completed, and decision makers can better understand the intelligent investment and financing decision making according to the decision tree, so that the decision making speed and efficiency are improved.
In other words, the decision tree is visually presented by means of the Echarts visual drawing tool. Through the visualization technology, the data can be converted into a graphical form, so that the data is more visual, easy to understand and analyze. This helps the user to discover information such as patterns, trends, outliers, etc. in the data, thereby better understanding the nature and meaning of the data.
Visualization techniques may help different users better understand and share data, thereby facilitating communication and collaboration between the financing teams. Through visualization techniques, users can express their own ideas and ideas more clearly, thereby better negotiating and making decisions. Visualization techniques may help users more accurately analyze and understand data to better make financing decisions. In addition, visualization techniques may also improve the efficiency of decision making because users may more quickly acquire and understand data, and thus make decisions more quickly.
Further, the investment and financing activity decision generation method further comprises the following steps:
respectively acquiring a plurality of corresponding characteristic information of a plurality of to-be-decided investment and financing activities, respectively inputting the corresponding plurality of characteristic information of each to-be-decided investment and financing activity into an investment and financing activity decision model, and obtaining a decision conclusion of each to-be-decided investment and financing activity;
judging the to-be-decided financing campaign which can be used as the optimal choice according to the decision conclusion of each to-be-decided financing campaign;
marking all decision branches in the decision tree, which are matched with the decision conclusion of each to-be-decided financing activity, respectively, and displaying the to-be-decided financing activity which can be used as the optimal selection and the decision tree with the mark through a visualization technology.
Another flow chart of the method for generating a financing activity decision provided by an embodiment of the present application can be seen in fig. 4.
By the method for generating the investment and financing activity decision, a decision tree marked with a decision branch related to the decision conclusion can be obtained while a decision object (activity) to be invested is input into a trained investment and financing activity decision model to obtain the corresponding decision conclusion; after the enterprise obtains the decision conclusion and the decision tree through the visualization technology, the decision conclusion and the decision tree are combined, and the path of the decision object to be invested in the decision tree can be clear, so that the thinking process that the decision conclusion is made by the investment and financing activity decision model is understood, how the decision conclusion is obtained by an algorithm in the investment and financing activity decision model is understood, and the transparency and the interpretability of the investment and financing activity decision model are improved, namely, the transparency and the interpretability of the investment and financing activity decision model by utilizing the artificial intelligence decision technology are improved.
Meanwhile, after a decision conclusion of an object to be invested is obtained by using the investment financing activity decision model, the decision tree can also help enterprises to correct and obtain the decision conclusion. For example, the investment and financing activity decision model determines that one of the decision objects to be invested is a high risk and high return, so that the conclusion of the decision object to be invested is a general concern. And because the investment risk that the enterprise can accept is higher, the enterprise can adjust the decision conclusion to be worth focusing, not just general. Thus, the decision tree can help the enterprise make a feedback or correction on the decision decisions that the investment and financing activity decision model gets.
Similarly, when a plurality of decision objects to be invested are input into the trained investment financing activity decision model, a decision conclusion of each decision object to be invested can be obtained, so that the decision tree marked with the decision conclusion about each decision object to be invested can be obtained, and the decision tree of which is the better choice can be known. The enterprise thus combines the more preferred conclusions with the decision tree, and is able to ascertain which path each decision object to be invested in belongs in the decision tree, and thus understand the thinking process of making the more preferred conclusions, and understand how to make the more preferred conclusions.
In other words, the artificial intelligence analysis corresponds to a black box, the internal thinking process of which is invisible and unintelligible, and the decision tree is shown as a supplementary description of the analysis result obtained by the artificial intelligence analysis. With reference to decision trees, enterprises can more easily understand ideas and thought processes of artificial intelligence analysis; meanwhile, according to the decision tree, the enterprise can correct or adjust artificial intelligence analysis to obtain decisions.
In summary, the method for generating the investment and financing activity decision provided by the embodiment of the application is taken as an intelligent analysis-based investment and financing management and decision method, and is a method for analyzing and deciding the investment and financing activity by utilizing an artificial intelligent technology. The method uses the visual technology, the decision tree and other methods to display the decision process and result of the algorithm by collecting, arranging and analyzing a large amount of investment and financing data and carrying out deep research on macroscopic and microscopic environments of markets, industries and enterprises, and enhances the transparency and the interpretability of the artificial intelligence decision technology, thereby providing comprehensive and accurate investment and financing decision support.
That is, in the method for generating the investment and financing activity decision provided by the embodiment of the application, in the process of investment and financing management decision, an intelligent decision method added with visualization and decision tree is used, so that a investment and financing manager can be helped to provide comprehensive and accurate investment and financing decision support:
1. the interpretability and transparency of the decision is enhanced. By using decision trees and visualization techniques, the user can clearly see what the basis of each decision is, helping people to better understand and trust decisions made by artificial intelligence.
2. The accuracy and the reliability of the decision are improved. Decision tree algorithms can classify a data set into multiple categories based on a set of features, thereby enabling classification or prediction. Through the visualization technology, a user can better understand the distribution and characteristics of the data, so that the accuracy and reliability of decision making are improved.
3. The speed and the efficiency of decision making are quickened. Decision tree algorithms and visualization techniques can help users quickly analyze and understand data, thereby speeding up decision making and efficiency.
4. And the decision adjustment and optimization are convenient. Through the visualization technology, a user can intuitively see the structure of the decision tree and the importance of each node, thereby facilitating the adjustment and optimization of decisions
It should be understood that, although the steps in the flowcharts of fig. 1 and 4 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least a portion of the steps in fig. 1 and 4 may include a plurality of steps or stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the execution of the steps or stages is not necessarily sequential, but may be performed in turn or alternately with at least a portion of the steps or stages in other steps or other steps.
In one embodiment, as shown in fig. 5, there is provided a financing activity decision generation apparatus comprising the following program modules:
a record information obtaining module 501, configured to obtain record information of a plurality of historical financing activities, where the record information includes a corresponding plurality of feature information and decision conclusion of each historical financing activity;
the model training and decision tree construction module 502 is configured to train to obtain a decision model of the investment and financing activity according to the record information of the plurality of historical investment and financing activities, and construct a corresponding decision tree;
the decision generation module 503 is configured to obtain a plurality of feature information of a to-be-decided investment and financing activity, input the plurality of feature information of the to-be-decided investment and financing activity into the investment and financing activity decision model, and obtain a decision conclusion of the to-be-decided investment and financing activity;
the visual display module 504 is configured to mark a decision branch in the decision tree that matches with a decision conclusion of the to-be-decided financing campaign, and display the decision conclusion of the to-be-decided financing campaign and the marked decision tree through a visual technology.
For a specific limitation of a financing decision making device, reference may be made to the limitation of a financing decision making method hereinabove, and the details are not repeated here. The above-mentioned various modules in a financing activity decision making apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 6. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a financing activity decision generation method provided by the above-described embodiments.
It will be appreciated by those skilled in the art that the structure shown in FIG. 6 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided, including a memory and a processor, the memory having stored therein a computer program, involving all or part of the flow of the methods of the embodiments described above.
In one embodiment, a computer readable storage medium having a computer program stored thereon is provided, involving all or part of the flow of the methods of the embodiments described above.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile memory may include Read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, or the like. Volatile memory can include Random access memory (Random AccessMemory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can take many forms, such as static random access memory (StaticRandomAccessMemory, SRAM) or dynamic random access memory (DynamicRandomAccessMemory, DRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (10)
1. A method for generating a financing activity decision, comprising:
s1, acquiring record information of a plurality of historical financing activities, wherein the record information comprises a plurality of corresponding characteristic information and decision conclusions of each historical financing activity;
s2, training to obtain a financing activity decision model according to the record information of the historical financing activities, and constructing a corresponding decision tree;
s3, obtaining a plurality of corresponding characteristic information of the to-be-decided investment and financing activity, and inputting the plurality of corresponding characteristic information of the to-be-decided investment and financing activity into the investment and financing activity decision model to obtain a decision conclusion of the to-be-decided investment and financing activity;
s4, marking a decision branch matched with the decision conclusion of the to-be-decided financing activity in the decision tree, and displaying the decision conclusion of the to-be-decided financing activity and the decision tree with the mark through a visualization technology.
2. The method of claim 1, wherein the plurality of characteristic information comprises a plurality of market factor information, a plurality of technical factor information, a plurality of financial factor information, a plurality of policy factor information, and a plurality of management factor information.
3. The method of claim 2, wherein the plurality of market factor information includes market size, market demand, and market competition, the plurality of technical factor information includes technical difficulty, technical maturity, and technical advantage, the plurality of financial factor information includes cost, expected return, return on investment for investment projects, the plurality of policy factor information includes government policies and laws and regulations, and the plurality of management factor information includes enterprise management level, team quality, and enterprise culture.
4. The method of claim 1, wherein step S1 further comprises:
judging whether the historical financing activities lack of characteristic information exist in the recorded information of the historical financing activities;
when the historical investment and financing activities with missing feature information are judged to exist, judging whether the number of the historical investment and financing activities with missing feature information exceeds a preset number threshold;
if the number of the historical investment and financing activities with the missing characteristic information is not more than a preset number threshold, deleting the recorded information of the historical investment and financing activities with the missing information;
if the number of the historical investment and financing activities of the missing feature information exceeds the preset number threshold, the value of the missing feature information is given to the common value or the average value of the corresponding feature.
5. The method of claim 1, wherein constructing the corresponding decision tree comprises:
selecting the characteristic information with the maximum information gain ratio from the recorded information of the plurality of historical financing activities as a root node of a decision tree by using a C4.5 algorithm;
and selecting optimal partition attributes layer by layer through the information gain ratio to construct the decision tree layer by layer.
6. The method of claim 1, wherein step S2 further comprises:
and pruning the decision tree after the decision tree is constructed.
7. A financing activity decision generation method according to claim 1, wherein in step S4 the decision tree with the labels is presented by means of an echart visual drawing tool.
8. The method of claim 7, wherein the presenting the marked decision tree by the echartis visual drawing tool comprises:
defining a JSON object containing data of said decision tree, said JSON object containing all nodes of said decision tree;
creating an empty canvas by using the Tree chart type in the Echarts visual drawing tool, and customizing the appearance of the canvas by setting the attribute of the canvas;
adding the data of the decision tree into the empty canvas by utilizing the JSON object;
defining the style of the node in the canvas by setting the attribute of the node;
the style of an edge in the canvas is defined by setting the properties of the edge.
9. The method of financing decision generation of claim 1, further comprising:
respectively acquiring a plurality of corresponding characteristic information of a plurality of to-be-decided investment and financing activities, respectively inputting the corresponding plurality of characteristic information of each to-be-decided investment and financing activity into the investment and financing activity decision model, and obtaining a decision conclusion of each to-be-decided investment and financing activity;
judging the to-be-decided financing campaign which can be used as the optimal choice according to the decision conclusion of each to-be-decided financing campaign;
marking all decision branches in the decision tree, which are matched with the decision conclusion of each to-be-decided financing activity, respectively, and displaying the to-be-decided financing activity which can be used as the optimal selection and the decision tree with the mark through a visualization technology.
10. A investment and financing activity decision-making device, comprising:
the system comprises a record information acquisition module, a decision making module and a decision making module, wherein the record information acquisition module is used for acquiring record information of a plurality of historical financing activities, and the record information comprises a plurality of corresponding characteristic information and decision making conclusions of each historical financing activity;
the model training and decision tree construction module is used for training to obtain a financing activity decision model according to the record information of the historical financing activities and constructing a corresponding decision tree;
the decision generation module is used for acquiring a plurality of corresponding characteristic information of the to-be-decided investment and financing activity, inputting the corresponding plurality of characteristic information of the to-be-decided investment and financing activity into the investment and financing activity decision model, and obtaining a decision conclusion of the to-be-decided investment and financing activity;
the visual display module is used for marking a decision branch matched with the decision conclusion of the to-be-decided financing activity in the decision tree, and displaying the decision conclusion of the to-be-decided financing activity and the decision tree with the mark through a visual technology.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310843148.0A CN116883175B (en) | 2023-07-10 | 2023-07-10 | Investment and financing activity decision generation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310843148.0A CN116883175B (en) | 2023-07-10 | 2023-07-10 | Investment and financing activity decision generation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116883175A true CN116883175A (en) | 2023-10-13 |
| CN116883175B CN116883175B (en) | 2024-07-02 |
Family
ID=88265640
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310843148.0A Active CN116883175B (en) | 2023-07-10 | 2023-07-10 | Investment and financing activity decision generation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116883175B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6278464B1 (en) * | 1997-03-07 | 2001-08-21 | Silicon Graphics, Inc. | Method, system, and computer program product for visualizing a decision-tree classifier |
| CN107292743A (en) * | 2017-06-07 | 2017-10-24 | 前海梧桐（深圳）数据有限公司 | The intelligent decision making method and its system invested and financed for enterprise |
| CN109389490A (en) * | 2018-09-26 | 2019-02-26 | 深圳壹账通智能科技有限公司 | Loan product matching process, device, computer equipment and storage medium |
| CN111242164A (en) * | 2019-12-27 | 2020-06-05 | 天津幸福生命科技有限公司 | Decision result determination method, device and equipment |
| CN111309852A (en) * | 2020-03-16 | 2020-06-19 | 青岛百洋智能科技股份有限公司 | Method, system, apparatus and storage medium for generating a visual decision tree (set) model |
| WO2021042843A1 (en) * | 2019-09-06 | 2021-03-11 | 平安科技（深圳）有限公司 | Alert information decision method and apparatus, computer device and storage medium |
| CN112699947A (en) * | 2020-12-30 | 2021-04-23 | 深圳前海微众银行股份有限公司 | Decision tree based prediction method, apparatus, device, medium, and program product |
| CN114048862A (en) * | 2021-11-10 | 2022-02-15 | 联想(北京)有限公司 | Model interpretation method and device |
| CN114418035A (en) * | 2022-03-25 | 2022-04-29 | 腾讯科技（深圳）有限公司 | Decision tree model generation method and data recommendation method based on decision tree model |
| CN115204411A (en) * | 2022-06-22 | 2022-10-18 | 杭州博盾习言科技有限公司 | Interpretation method, system, device and medium based on decision tree and LIME model |
| CN115409272A (en) * | 2022-09-02 | 2022-11-29 | 安徽维德工业自动化有限公司 | Dynamic flexible optimization method of production process based on regression decision tree |
| CN115809928A (en) * | 2022-12-15 | 2023-03-17 | 蚂蚁区块链科技(上海)有限公司 | Project investment decision-making method and system |
| CN116384480A (en) * | 2023-04-04 | 2023-07-04 | 南京大学深圳研究院 | Deep reinforcement learning decision interpretation system |
-
2023
- 2023-07-10 CN CN202310843148.0A patent/CN116883175B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6278464B1 (en) * | 1997-03-07 | 2001-08-21 | Silicon Graphics, Inc. | Method, system, and computer program product for visualizing a decision-tree classifier |
| CN107292743A (en) * | 2017-06-07 | 2017-10-24 | 前海梧桐（深圳）数据有限公司 | The intelligent decision making method and its system invested and financed for enterprise |
| CN109389490A (en) * | 2018-09-26 | 2019-02-26 | 深圳壹账通智能科技有限公司 | Loan product matching process, device, computer equipment and storage medium |
| WO2021042843A1 (en) * | 2019-09-06 | 2021-03-11 | 平安科技（深圳）有限公司 | Alert information decision method and apparatus, computer device and storage medium |
| CN111242164A (en) * | 2019-12-27 | 2020-06-05 | 天津幸福生命科技有限公司 | Decision result determination method, device and equipment |
| CN111309852A (en) * | 2020-03-16 | 2020-06-19 | 青岛百洋智能科技股份有限公司 | Method, system, apparatus and storage medium for generating a visual decision tree (set) model |
| CN112699947A (en) * | 2020-12-30 | 2021-04-23 | 深圳前海微众银行股份有限公司 | Decision tree based prediction method, apparatus, device, medium, and program product |
| CN114048862A (en) * | 2021-11-10 | 2022-02-15 | 联想(北京)有限公司 | Model interpretation method and device |
| CN114418035A (en) * | 2022-03-25 | 2022-04-29 | 腾讯科技（深圳）有限公司 | Decision tree model generation method and data recommendation method based on decision tree model |
| CN115204411A (en) * | 2022-06-22 | 2022-10-18 | 杭州博盾习言科技有限公司 | Interpretation method, system, device and medium based on decision tree and LIME model |
| CN115409272A (en) * | 2022-09-02 | 2022-11-29 | 安徽维德工业自动化有限公司 | Dynamic flexible optimization method of production process based on regression decision tree |
| CN115809928A (en) * | 2022-12-15 | 2023-03-17 | 蚂蚁区块链科技(上海)有限公司 | Project investment decision-making method and system |
| CN116384480A (en) * | 2023-04-04 | 2023-07-04 | 南京大学深圳研究院 | Deep reinforcement learning decision interpretation system |
Non-Patent Citations (4)
| Title |
|---|
| AI科技大本营: "机器学习的可解释性：黑盒vs白盒（内附开源学习书）|Deep Reading", pages 1 - 4, Retrieved from the Internet <URL:https://zhuanlan.zhihu.com/p/61172671> * |
| 岳意定 等: "基于人工神经网络的投资项目效益评估模型", 《矿治工程》, vol. 27, no. 6, pages 86 - 89 * |
| 岳意定;刘丹;曾之明;: "基于人工神经网络的投资项目效益评估模型", 矿冶工程, no. 06, 15 December 2007 (2007-12-15), pages 86 - 89 * |
| 数据分析: "全面：手把手教你决策树可视化（附链接&代码）", pages 1 - 24, Retrieved from the Internet <URL:https://www.sohu.com/a/370408120_197042> * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116883175B (en) | 2024-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10417523B2 (en) | Dimension grouping and reduction for model generation, testing, and documentation | |
| Amar et al. | Best paper: A knowledge task-based framework for design and evaluation of information visualizations | |
| US10565750B2 (en) | Intelligent visualization munging | |
| Sedlmair et al. | Visual parameter space analysis: A conceptual framework | |
| Amar et al. | Knowledge precepts for design and evaluation of information visualizations | |
| Provost et al. | Data science and its relationship to big data and data-driven decision making | |
| CN105718490A (en) | Method and device for updating classifying model | |
| Kulkarni et al. | Evolve systems using incremental clustering approach | |
| CN109933783A (en) | A kind of essence of a contract method of non-performing asset operation field | |
| Silver et al. | A Markov model of urban evolution: Neighbourhood change as a complex process | |
| Nandi et al. | Data Science fundamentals and practical approaches: understand why data science is the next | |
| Collaris et al. | Strategyatlas: Strategy analysis for machine learning interpretability | |
| Goerigk et al. | A framework for inherently interpretable optimization models | |
| Sokol et al. | Interpretable representations in explainable AI: from theory to practice | |
| Sunahara et al. | Universal productivity patterns in research careers | |
| Mujthaba et al. | Data Science Techniques, Tools and Predictions | |
| CN116883175B (en) | Investment and financing activity decision generation method and device | |
| Rhyn et al. | Patterns of Data-Driven Decision-Making: How Decision-Makers Leverage Crowdsourced Data. | |
| Saleh et al. | Design of decision support system for loans based on data mining techniques | |
| Ozen et al. | The Role of Artificial Intelligence (AI) in Improving Product Development Efficiency | |
| CN113496222B (en) | Mode identification method and device based on label-free chart data and computer equipment | |
| Azuar et al. | Interactive Dashboard For Tracking System Dashboard Using Power Bi | |
| CN118379033B (en) | Contract relation management method and system based on graph data | |
| Abd Rahman et al. | Genetic Programming based Automated Machine Learning in Classifying ESG Performances | |
| LUCHIAN et al. | UNLEASHING THE POWER OF DATA: THE SIGNIFICANCE OF ANALYSIS AND PREDICTION |
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 |