A project management supplier decision analysis method based on integrated algorithm

Abstract

A project management supplier decision analysis method based on an integrated algorithm includes: collecting supplier attribute data that affects decision factors through a project management platform; preprocessing the collected data; reading attribute data information to determine whether the attribute information is continuous, and if the attribute information is continuous, discretizing the continuous value by level; storing the discretized data to obtain a training set, and then dividing the training set, processing its subsets using an improved C4.5 algorithm to obtain an initial decision tree; using the obtained initial decision tree, using an improved integrated algorithm, generating a final decision tree to make a decision on the supplier. The present invention directly uses a genetic algorithm for the optimization of decision trees. Since the decision tree generated by the C4.5 algorithm has a high classification accuracy, using the C4.5 algorithm to generate an initial decision tree population can effectively avoid the genetic algorithm from blindly searching at the beginning, solving the problem of the prior art that the C4.5 balanced decision tree algorithm is prone to overfitting.

Description

Technical Field

[0001] The present invention relates to the field of data mining classification, and in particular to a project management supplier decision analysis method based on integrated algorithm. Background Art

[0002] At present, most decision analysis methods only consider using the original C4.5 algorithm, which cannot solve the overfitting problem of the C4.5 balanced decision tree algorithm. Or using its improved algorithm to adjust the calculation method of the attribute gain rate to find out the reasons that affect the decision, or optimizing the original data set in the data processing stage, it cannot significantly improve the algorithm calculation time and classification accuracy. Summary of the invention

[0003] In view of the above problems, the present invention is proposed to provide a project management supplier decision analysis method based on an integrated algorithm that overcomes the above problems or at least partially solves the above problems.

[0004] In order to solve the above technical problems, the embodiments of the present application disclose the following technical solutions:

[0005] The present invention discloses a project management supplier decision analysis method based on an integrated algorithm, comprising:

[0006] S100. Collect supplier attribute data that influences decision-making factors through the project management platform;

[0007] S200. Preprocessing the collected data;

[0008] S300. Read the attribute data information, determine whether the attribute information is continuous, and if the attribute information is continuous, discretize the continuous value according to the level;

[0009] S400. The discretized data is stored to obtain a training set, and then the training set is divided, and its subsets are processed using an improved C4.5 algorithm to obtain an initial decision tree;

[0010] S500. Generate a final decision tree using the obtained initial decision tree and improve the integrated algorithm to make a decision on the supplier.

[0011] Furthermore, in S100, there are five types of supplier attribute data that affect decision-making factors, including: supply quality, delivery time, service quality, supply price, and credit rating.

[0012] Furthermore, it is characterized in that, in S200, preprocessing the data at least includes: null value processing and data set integration.

[0013] Furthermore, it is characterized in that, in S300, the supply price and credit rating are continuous data, which need to be divided into intervals and then discretized by taking the interval levels; the supply quality, delivery time and service quality are discrete data and do not need to be processed.

[0014] Furthermore, the improved C4.5 algorithm is used for processing. The specific method is as follows: first, the information entropy is calculated through the attribute data, and the balance factor is introduced to obtain the attribute information gain rate. The results of the information gain rates of all attributes are obtained, and the attribute with the largest information gain rate is selected as the split attribute of the current node. Other attributes will continue to be recursively calculated. As the attributes are gradually calculated, the information gain rate will become smaller and smaller. The attributes with relatively large information gain rates are selected as classification attributes until the final decision tree is generated.

[0015] Further, in S500, an improved integrated algorithm is used for processing, including:

[0016] S501. Generate an initial decision tree using the optimized C4.5 algorithm, and generate a corresponding rule set according to the data rules given by the decision tree;

[0017] S502. Encode the rule set to generate an initial population;

[0018] S503. Determine the fitness function, and use the three operators of genetic operation to repeatedly perform genetic operation on the individuals in the initial population;

[0019] S504. Obtaining better individuals through the calculation of the fitness function and preserving them generation by generation until the termination condition is met;

[0020] S505. The rule set with the largest fitness function value is retained and decoded, thereby obtaining a relatively better decision tree model.

[0021] Furthermore, three operators of genetic operation are used, including: selection operator, crossover operator and mutation operator.

[0022] The beneficial effects of the above technical solution provided by the embodiment of the present invention include at least:

[0023] The present invention discloses a project management supplier decision analysis method based on an integrated algorithm, comprising: collecting supplier attribute data that affects decision factors through a project management platform; preprocessing the collected data; reading attribute data information, judging whether the attribute information is continuous, and if the attribute information is continuous, discretizing the continuous value by level; storing the discretized data to obtain a training set, and then dividing the training set, processing its subsets using an improved C4.5 algorithm to obtain an initial decision tree; using the obtained initial decision tree, using an improved integrated algorithm, generating a final decision tree to make a decision on the supplier. The present invention directly uses a genetic algorithm for the optimization of a decision tree. Since the decision tree generated by the C4.5 algorithm has a high classification accuracy, the C4.5 algorithm is first used to generate an initial decision tree population, which can effectively avoid the genetic algorithm from blindly searching at the beginning, and solves the problem of the prior art that the C4.5 balanced decision tree algorithm is very easy to overfit.

[0024] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0026] Figure 1 This is a flowchart of a project management supplier decision analysis method based on an integrated algorithm in Example 1 of the present invention. DETAILED DESCRIPTION

[0027] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0028] In order to solve the problems existing in the prior art, an embodiment of the present invention provides a project management supplier decision analysis method based on an integrated algorithm.

[0029] Example 1

[0030] The present invention discloses a project management supplier decision analysis method based on an integrated algorithm, such as Figure 1 ,include:

[0031] S100. Collect the supplier attribute data of the factors affecting the decision through the project management platform; in this embodiment, the results of the decision of the procurement supplier have four values: "excellent", "good", "medium" and "poor". They are represented by A, B, C and D respectively. There are five factors affecting the decision, including supply quality (Qu), delivery time (Ti), service (Se), supply price (Pr) and credit (Cr).

[0032] S200. Preprocess the collected data; specifically, the data extracted from the system platform is often incomplete, so before data mining, data preprocessing is required, including null value processing, data set integration, etc.

[0033] S300. Read the attribute data information and determine whether the attribute information is continuous. If the attribute information is continuous, the continuous value is discretized according to the level. In this embodiment, the discrete data of supply quality, delivery time and service do not need to be processed. The continuous data of supply price and credit rating need to be divided into intervals and then discretized according to the interval level.

[0034] S400. The discretized data is stored to obtain a training set, and then the training set is divided, and its subsets are processed using the improved C4.5 algorithm to obtain an initial decision tree. Specifically, the improved C4.5 algorithm is used for processing, and the specific method is as follows: first, the information entropy is calculated through the attribute data, and the balance factor is introduced to obtain the attribute information gain rate, and the result of the information gain rate of all attributes is obtained. The attribute with the largest information gain rate is selected as the splitting attribute of the current node, and other attributes will continue to be recursively calculated. As the attributes are gradually calculated, the information gain rate will become smaller and smaller, and the attribute with a relatively large information gain rate is selected as the classification attribute until the final decision tree is generated.

[0035] Specifically, information entropy is calculated. Information entropy represents the value of information. The lower the information entropy, the greater the probability of the information appearing. The calculation formula of information entropy is:

[0036]

[0037]

[0038] The logarithm is usually taken as base 2, X represents a random variable, and the corresponding variable is the set of all possible outputs, which is defined as a symbol set. The number of symbol sets is n and the number of types is k. i represents the output probability function. E(X i ) represents attribute expectation, γ i To balance the factors, the greater the uncertainty of the variable, the greater the entropy, and the more information is needed.

[0039] Taking the attribute supply quality Qu as an example, the information gain of the sample set after the split according to the attribute Qu is calculated. The partition information entropy of the attribute Qu partition X is: (where |X j |, |X| represents the number of samples included)

[0040]

[0041] Gain(Qu)=H(X)-H Qu (X)

[0042] The initial information amount of the split of attribute supply quality Qu

[0043]

[0044] The information gain rate of the sample set after the attribute supply quality Qu is split

[0045]

[0046] After the improved C4.5 algorithm is executed, the information gain rate of all attributes is obtained, and the attribute with the largest information gain rate is selected as the splitting attribute of the current node. Other attributes will continue to be recursively calculated. As the attributes are gradually calculated, the information gain rate will become smaller and smaller, and the attributes with relatively large information gain rates are selected as classification attributes until the final decision tree is generated.

[0047] S500. Generate a final decision tree using the obtained initial decision tree and improve the integrated algorithm to make a decision on the supplier.

[0048] Specifically, in S500, an improved integrated algorithm is used for processing, including:

[0049] S501. Generate an initial decision tree using the optimized C4.5 algorithm, and generate a corresponding rule set according to the data rules given by the decision tree;

[0050] S502. Encode the rule set and generate an initial population; construct corresponding decision trees by selecting different subsets in the data set, and convert them into corresponding rule sets for chromosome binary encoding, using numerical values ​​of 0 or 1 to represent chromosomes in an arrangement order such as 1010010.

[0051] S503. Determine the fitness function, and use the three operators of genetic operation to repeatedly perform genetic operation on the individuals in the initial population;

[0052] In this embodiment, the three operators of genetic operation specifically include: a selection operator, a crossover operator, and a mutation operator.

[0053] ①Select operator

[0054] Genetic algorithms use selection operators to represent the reproduction process of individuals in a population. The fitness function is used as the selection criterion. Individuals with high fitness values ​​have strong adaptability to the environment and are more likely to pass on genes to the next generation.

[0055] ② Crossover operator

[0056] The crossover operator plays a role in promoting the diversity of individuals in the population in the genetic algorithm. First, the mutation point or mutation interval is randomly determined. Then, the individuals in the population are allowed to cross over and exchange with each other according to a certain probability (crossover probability), resulting in a new gene arrangement order, that is, the emergence of new individuals, reflecting the global search ability of the genetic algorithm. Commonly used crossover methods include single-point crossover, two-point crossover, multi-point crossover, and uniform crossover, etc.

[0057] ③ Mutation operator

[0058] The mutation operator operates on independent individuals, and changes the genes at a certain point or interval in the chromosome of a single individual according to a certain probability (mutation probability), thereby generating new individuals. In order to ensure that good genes are not lost and the overall stability of the population, mutation operators are often performed at a relatively small probability.

[0059] S504. Obtaining better individuals through the calculation of the fitness function and preserving them generation by generation until the termination condition is met;

[0060] S505. The rule set with the largest fitness function value is retained and decoded, thereby obtaining a relatively better decision tree model.

[0061] The present embodiment discloses a project management supplier decision analysis method based on an integrated algorithm, including: collecting supplier attribute data that affects decision factors through a project management platform; preprocessing the collected data; reading attribute data information to determine whether the attribute information is continuous, and if the attribute information is continuous, discretizing the continuous value by level; storing the discretized data to obtain a training set, and then dividing the training set, processing its subsets using an improved C4.5 algorithm to obtain an initial decision tree; using the obtained initial decision tree, using an improved integrated algorithm, generating a final decision tree to make a decision on the supplier. The present invention directly uses a genetic algorithm for the optimization of decision trees. Since the decision tree generated by the C4.5 algorithm has a high classification accuracy, the C4.5 algorithm is first used to generate an initial decision tree population, which can effectively avoid the genetic algorithm from blindly searching at the beginning, and solves the problem of the prior art that the C4.5 balanced decision tree algorithm is very easy to overfit.

[0062] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process can be rearranged without departing from the scope of protection of the present disclosure. The attached method claims present the elements of the various steps in an exemplary order and are not intended to be limited to the specific order or hierarchy described.

[0063] In the above detailed description, various features are grouped together in a single embodiment to simplify the disclosure. This method of disclosure should not be interpreted as reflecting an intention that the embodiments of the claimed subject matter require more features than are clearly stated in each claim. On the contrary, as reflected in the appended claims, the invention is in a state of having less than all the features of the disclosed individual embodiments. Therefore, the appended claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.

[0064] Those skilled in the art will also appreciate that the various illustrative logic blocks, modules, circuits, and algorithmic steps described in conjunction with the embodiments herein can all be implemented as electronic hardware, computer software, or a combination thereof. In order to clearly illustrate the interchangeability between hardware and software, various illustrative components, blocks, modules, circuits, and steps are generally described above around their functions. Whether such functions are implemented as hardware or software depends on specific applications and the design constraints imposed on the entire system. A skilled person can implement the described functions in an alternative manner for each specific application, but such implementation decisions should not be interpreted as departing from the scope of protection of the present disclosure.

[0065] The steps of the method or algorithm described in conjunction with the embodiments herein may be directly embodied as hardware, a software module executed by a processor, or a combination thereof. The software module may be located in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a mobile disk, a CD-ROM, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Of course, the storage medium may also be an integral part of the processor. The processor and the storage medium may be located in an ASIC. The ASIC may be located in a user terminal. Of course, the processor and the storage medium may also be present in a user terminal as discrete components.

[0066] For software implementation, the techniques described in this application can be implemented with modules (e.g., procedures, functions, etc.) that perform the functions described in this application. These software codes can be stored in a memory unit and executed by a processor. The memory unit can be implemented within the processor or outside the processor. In the latter case, it is coupled to the processor in a communication manner via various means, which are well known in the art.

[0067] The above description includes examples of one or more embodiments. Of course, it is impossible to describe all possible combinations of components or methods for the purpose of describing the above embodiments, but it should be recognized by those skilled in the art that the various embodiments may be further combined and arranged. Therefore, the embodiments described herein are intended to cover all such changes, modifications and variations that fall within the scope of protection of the appended claims. In addition, with respect to the term "comprising" used in the specification or claims, the word is covered in a manner similar to the term "including", just as "including," is explained as a transitional word in the claims. In addition, any term "or" used in the specification of the claims is intended to mean "non-exclusive or".

Claims

1. A project management supplier decision analysis method based on an integrated algorithm, characterized in that: include: S100. Collect supplier attribute data that influences decision-making factors through the project management platform; In S100, there are five types of supplier attribute data that affect decision-making factors, including: supply quality, delivery time, service quality, supply price, and credit rating; S200. Preprocessing the collected data; S300. Read the attribute data information, determine whether the attribute information is continuous, and if the attribute information is continuous, discretize the continuous value according to the level; S400. The discretized data is stored to obtain a training set, and then the training set is divided, and its subsets are processed using the improved C4.5 algorithm to obtain an initial decision tree; the improved C4.5 algorithm is used for processing, and the specific method is as follows: first, the information entropy is calculated through the attribute data, and the balance factor is introduced to obtain the attribute information gain rate, and the results of the information gain rates of all attributes are obtained, and the attribute with the largest information gain rate is selected as the splitting attribute of the current node, and other attributes will continue to be recursively calculated. As the attributes are gradually calculated, the information gain rate will become smaller and smaller, and the attribute with a relatively large information gain rate is selected as the classification attribute until the final decision tree is generated; Specifically, information entropy is calculated. Information entropy represents the value of information. The lower the information entropy, the greater the probability of the information appearing. The calculation formula of information entropy is: The logarithm is usually taken as base 2, X represents a random variable, and the corresponding one is the set of all possible outputs, which is defined as a symbol set. The number of symbol sets is n, the types are k, and p i represents the output probability function; EX i represents attribute expectation, γ i To balance the factors, the greater the uncertainty of the variables, the greater the entropy, and the more information is needed; Taking the attribute supply quality Qu as an example, the information gain of the sample set after the split according to the attribute Qu is calculated, and the partition information entropy of the attribute Qu partition X is: GainQu=HX-H Qu (X) The initial information amount of the split of the attribute supply quality Qu: The information gain rate of the sample set after the attribute supply quality Qu is split: Among them, |X j |, |X| represent the number of samples included; S500. Generate a final decision tree by using the improved integrated algorithm through the obtained initial decision tree to make a decision on the supplier; In S500, the improved integrated algorithm is used for processing, including: S501. Generate an initial decision tree using the optimized C4.5 algorithm, and generate a corresponding rule set according to the data rules given by the decision tree; S502. Encode the rule set to generate an initial population; S503. Determine the fitness function, and use the three operators of genetic operation to repeatedly perform genetic operation on the individuals in the initial population; S504. Obtaining better individuals through the calculation of the fitness function and preserving them generation by generation until the termination condition is met; S505. The rule set with the largest fitness function value is retained and decoded, thereby obtaining a relatively better decision tree model.

2. A project management supplier decision analysis method based on integrated algorithm as claimed in claim 1, characterized in that: In S200, the data is preprocessed, which at least includes: null value processing and data set integration.

3. A project management supplier decision analysis method based on integrated algorithm as claimed in claim 2, characterized in that: In S300, the supply price and credit rating are continuous data, which need to be divided into intervals and then discretized by taking the interval levels; the supply quality, delivery time, and service quality are discrete data and do not need to be processed.

4. A project management supplier decision analysis method based on integrated algorithm as claimed in claim 1, characterized in that: The three operators of genetic operation include selection operator, crossover operator and mutation operator.

Images