Potential competitor analysis method, system and device and storage medium

Abstract

The invention provides a potential competitor analysis method, system and device and a storage medium. The potential competitor analysis method comprises the steps of inputting section data, and screening historical section data based on discrete features of the section data to obtain candidate similar sections; calculating the text similarity between the bidding section and the candidate similar sections and the numerical deviation of the estimated price of the bidding contract, comparing the text similarity and the numerical deviation with a threshold value, and screening the candidate similar sections to obtain a final similar section; inputting the bid-winning price and the number of bidding units corresponding to the final similar sections into a logistic regression model for prediction, generating a predicted floating rate based on the bid-winning price corresponding to the final similar sections, and obtaining the predicted bid-winning price and the number of bidding units based on the predicted floating rate; and displaying a prediction result of the bidding probability based on the final similar sections, and optimizing prediction column data and counting quotation distribution. According to the method, the problems of high time consumption, low efficiency and inaccurate analysis in analysis of potential competitors are solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a potential competitor analysis method, system, device and storage medium. Background Art

[0002] During the bidding process, the bidding unit will fully understand the contents of the bidding documents and find out the key information of the project, which may include project description, project qualification requirements, performance requirements, bid evaluation methods, etc. According to the requirements of the bidding documents, potential competitors who may bid and the possible price range of the project will be analyzed to formulate corresponding bidding strategies.

[0003] Due to the large number of bidding units in various regions, bidding units mainly rely on manual analysis to analyze potential competitors. Therefore, bidders not only spend more time on data collection, but also the manual analysis of potential competitors is inefficient and the analysis results are inaccurate and prone to errors, resulting in the inability to formulate corresponding bidding strategies and thus affecting the winning of the bid. Summary of the invention

[0004] In view of this, the embodiments of the present application are committed to providing a potential competitor analysis method, system, device and storage medium, which are used to solve the problem that bidding units spend a lot of time on data collection during the bidding process, and the manual analysis of potential competitors is inefficient and the analysis results are inaccurate, resulting in the inability to formulate corresponding bidding strategies and thus affecting the winning of the bid.

[0005] According to a first aspect of the present invention, the present invention provides a potential competitor analysis method, comprising:

[0006] Obtain historical bidding section data and pre-process the historical bidding section data;

[0007] Input the bidding section data to be bid, and screen the pre-processed historical bidding section data based on the discrete features of the bidding section data to be bid, so as to obtain candidate similar bidding sections;

[0008] Calculate the text similarity between the bidding section and the candidate similar sections and the numerical deviation of the estimated price of the bidding contract, compare the calculated text similarity and numerical deviation with the preset threshold, and re-screen the candidate similar sections to obtain the final similar sections;

[0009] The winning bid price and the number of bidding units corresponding to the final similar bidding sections are input into the pre-trained logistic regression model for prediction, and a predicted discount rate is generated based on the winning bid price corresponding to the final similar bidding sections, and the predicted winning bid price and the number of bidding units are obtained based on the predicted discount rate;

[0010] Based on the prediction results of the bidding probability of the final similar bidding sections, the prediction column data is optimized and the quotation distribution is statistically analyzed.

[0011] Preferably, the obtaining of historical bid section data and preprocessing of the historical bid section data includes:

[0012] The contract estimated price and the bid winning price in the historical bidding data are pre-processed to unify the numerical units.

[0013] Preferably, the inputting of the bid section data to be bid, screening the pre-processed historical bid section data based on the discrete features of the bid section data to be bid, and obtaining candidate similar bid sections comprises:

[0014] The historical bidding section data is screened according to various features of the bidding section data to be bid, wherein the various features include text features, discrete features, time series features and numerical features.

[0015] Preferably, the screening of historical bidding section data according to multiple characteristics of the bidding section data to be bid includes:

[0016] Through discrete feature screening, the historical bidding section data is preliminarily screened using the discrete features of region and project transaction classification. The historical bidding section data that pass the screening are candidate similar bidding sections.

[0017] Preferably, the calculating of the text similarity between the bid section and the candidate similar sections and the numerical deviation of the bid contract estimated price, comparing the calculated text similarity and numerical deviation with a preset threshold, and re-screening the candidate similar sections to obtain the final similar sections comprises:

[0018] Calculate the text similarity between the bidding section and all candidate similar sections, compare the calculation result with the preset threshold, and use the text similarity higher than the preset threshold as one of the screening items;

[0019] Calculate the numerical deviation between the estimated contract price of the bidding section and the estimated contract prices of all candidate similar bidding sections, then compare the calculated result with a preset threshold, and use the numerical deviation less than the preset threshold as another screening item;

[0020] The candidate similar bid sections that meet all the above screening items at the same time are the final similar bid sections.

[0021] Preferably, the winning bid price and the number of bidding units corresponding to the final similar bidding sections are input into a pre-trained logistic regression model for prediction, a predicted downward floating rate is generated based on the winning bid price corresponding to the final similar bidding sections, and the predicted winning bid price and the number of bidding units are obtained based on the predicted downward floating rate, including:

[0022] Input the winning bid prices and number of bid units corresponding to all final similar bid sections into the pre-trained logistic regression model for prediction;

[0023] Use the following formula to calculate the predicted downward floating rate of each final similar bid section, count the downward floating rates of all final similar bid sections, and take the average of all downward floating rates as the predicted downward floating rate:

[0024] (Winning bid price - contract estimated price) / contract estimated price;

[0025] The predicted winning bid price is obtained by calculating contract estimated price * predicted discount rate = predicted winning bid price.

[0026] Preferably, the method of displaying the prediction results of the bidding probability based on the final similar bidding sections, optimizing the prediction column data and statistically analyzing the quotation distribution includes:

[0027] The bidding data is associated through the Guid of the final similar bidding sections, and grouped according to the unit name in the discrete feature;

[0028] According to the bidding times and bidding section times of each bidding unit in the historical bidding data, the bidding probability of each bidding unit is calculated as follows:

[0029]

[0030] Sort the bidders in each group by bid probability from high to low, and select the bidders with high bid probability according to the number of bidders predicted previously. If the amount of valid data is insufficient, update the number of bidders to the corresponding number of bidders;

[0031] For each bidding unit, calculate (quoted price - contract estimated price) / contract estimated price in each bidding section as the discount rate, obtain the average as the discount rate of the corresponding company's quotation, calculate the contract estimated price * (1 + discount rate) as the corresponding company's quotation, and count the quotation distribution of each bidding unit as the final output.

[0032] According to a second aspect of the present invention, there is also provided a potential competitor analysis system, comprising:

[0033] A preprocessing module, used to obtain historical bid section data and preprocess the historical bid section data;

[0034] A screening module is used to input the bidding section data to be bid, screen the pre-processed historical bidding section data based on the discrete features of the bidding section data to be bid, and obtain candidate similar bidding sections;

[0035] The prediction module is used to calculate the text similarity between the bidding section and the candidate similar sections and the numerical deviation of the bidding contract estimated price, compare the calculated text similarity and numerical deviation with the preset threshold, and re-screen the candidate similar sections to obtain the final similar sections;

[0036] The probability module is used to input the winning bid price and the number of bidding units corresponding to the final similar bidding sections into the pre-trained logistic regression model for prediction, generate a predicted downward floating rate based on the winning bid price corresponding to the final similar bidding sections, and obtain the predicted winning bid price and the number of bidding units based on the predicted downward floating rate;

[0037] The quotation module is used to display the prediction results of bidding probability based on the final similar bidding sections, optimize the prediction column data and count the quotation distribution.

[0038] According to a third aspect of the present invention, an electronic device is also provided, comprising a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the potential competitor analysis method.

[0039] According to a fourth aspect of the present invention, a computer-readable storage medium is also provided, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the potential competitor analysis method.

[0040] The effects provided in the content of the invention are only the effects of the embodiments, not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:

[0041] According to the first aspect of the present invention, the present invention aims to improve the efficiency of the bidder's analysis of competitors and reduce the time and error of manual analysis. The method pre-processes the historical bidding data, screens out the biddings similar to the current bidding project, and screens their text similarity and contract estimated price deviation to further narrow the scope of similar biddings. Finally, the screened similar bidding data is input into a logistic regression or linear regression model to predict the bid price, discount rate and bid probability, and obtain the quotation distribution of the bidding unit. This solution automates the analysis process of competitors, provides more accurate support for bidding strategies through data screening and prediction models, thereby helping bidders to formulate better bidding strategies and effectively improve the probability of winning the bid. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A schematic flow chart of a potential competitor analysis method according to an embodiment of the present invention is shown;

[0043] Figure 2 A schematic flow chart of a potential competitor analysis system according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0044] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0045] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0046] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0047] Figure 1 FIG. 4 is a schematic flow chart of a potential competitor analysis method according to an embodiment of the present invention. Figure 1 As shown, the present invention provides a potential competitor analysis method, comprising:

[0048] Step S100, obtaining historical bid section data and preprocessing the historical bid section data;

[0049] In step S100, the historical bidding data is preprocessed. Specifically, the historical bidding data on the market is first collected, and then the contract estimated price and the bid winning price in the historical bidding data are preprocessed to unify the numerical units.

[0050] Step 200: input the bidding section data to be bid, and screen the pre-processed historical bidding section data based on the discrete features of the bidding section data to be bid, to obtain candidate similar bidding sections.

[0051] In step S200, the historical bidding section data is screened according to various features of the bidding section data (i.e., the bidding section) to be bid, wherein the various features include text features, discrete features, time series features, and numerical features, etc. In the present application, discrete feature screening is performed to perform preliminary screening on the historical bidding section data using discrete features such as region, project transaction classification, and the historical bidding section data after screening is the candidate similar bidding section.

[0052] Step 300, calculate the text similarity between the bid section and the candidate similar sections and the numerical deviation of the bid contract estimated price, compare the calculated text similarity and numerical deviation with the preset threshold, and re-screen the candidate similar sections to obtain the final similar sections.

[0053] In step 300, the text similarity between the bid section and all candidate similar sections is calculated, and the calculation result is compared with a preset threshold, and the text similarity higher than the preset threshold is used as one of the screening items. At the same time, the numerical deviation between the estimated contract price of the bid section and the estimated contract price of all candidate similar sections is calculated, and the calculation result is compared with a preset threshold, and the numerical deviation less than the preset threshold is used as another screening item. Finally, the candidate similar section that satisfies all the above screening items at the same time is the final similar section.

[0054] It should be noted that the calculation of the text similarity between the bid section and all candidate similar sections in this application is a prior art and will not be described in detail here.

[0055] Step 400: Input the winning bid price and the number of bidding units corresponding to the final similar bidding section into a pre-trained logistic regression model for prediction, generate a predicted discount rate based on the winning bid price corresponding to the final similar bidding section, and obtain the predicted winning bid price and the number of bidding units based on the predicted discount rate.

[0056] In step 400, the winning bid prices and the number of bidders corresponding to all final similar bid sections are input into the pre-trained logistic regression model for prediction, and the following formula is used: (winning bid price - contract estimated price) / contract estimated price; the predicted downward floating rate of each final similar bid section is calculated, the downward floating rates of all final similar bid sections are counted, and the average of all downward floating rates is used as the predicted downward floating rate. Finally, the predicted winning bid price is obtained by contract estimated price * predicted downward floating rate = predicted winning bid price.

[0057] Optionally, in the present application, the winning bid price and the number of bidding units are predicted through an existing logistic regression or linear regression model.

[0058] Step 500: Based on the prediction results of the bidding probability displayed for the final similar bidding sections, the prediction column data is optimized and the quotation distribution is statistically analyzed.

[0059] In step 500, the bidding data is first associated with the Guid of the final similar bidding section, and grouped according to the unit name in the discrete feature; then the bidding probability of each bidding unit is calculated according to the bidding number and bidding section number of each bidding unit in the historical bidding data as follows:

[0060]

[0061] Sort the bidders in each group by bid probability from high to low, and select the bidders with high bid probability according to the number of bidders predicted before. If the amount of valid data is insufficient, update the number of bidders to the corresponding number of bidders. For each bidder, calculate (quotation - contract estimate) / contract estimate in each bid section as the discount rate, obtain the mean as the discount rate of the corresponding company's quotation, calculate the contract estimate * (1 + discount rate) as the quotation of the corresponding company, and finally count the quotation distribution of each bidder as the final output.

[0062] In summary, this application aims to improve the efficiency of bidders' analysis of competitors and reduce the time and errors of manual analysis. This method pre-processes historical bidding data, screens out biddings similar to the current bidding project, and screens their text similarity and contract estimate price deviations to further narrow the scope of similar biddings. Finally, the screened similar bidding data is input into a logistic regression or linear regression model to predict the bid price, discount rate and bid probability, and obtain the quotation distribution of the bidding unit. This solution automates the analysis process of competitors, provides more accurate support for bidding strategies through data screening and prediction models, and thus helps bidders formulate better bidding strategies and effectively improve the probability of winning the bid.

[0063] like Figure 2 As shown, a potential competitor analysis system is provided, comprising:

[0064] A preprocessing module 810 is used to obtain historical bid section data and preprocess the historical bid section data;

[0065] The screening module 820 is used to input the bidding section data to be bid, and screen the pre-processed historical bidding section data based on the discrete features of the bidding section data to be bid, so as to obtain candidate similar bidding sections;

[0066] Prediction module 830, used to calculate the text similarity between the bid section and the candidate similar sections and the numerical deviation of the bid contract estimated price, compare the calculated text similarity and numerical deviation with a preset threshold, and re-screen the candidate similar sections to obtain the final similar sections;

[0067] The probability module 840 is used to input the winning bid price and the number of bidding units corresponding to the final similar bidding section into the pre-trained logistic regression model for prediction, generate a predicted downward floating rate based on the winning bid price corresponding to the final similar bidding section, and obtain the predicted winning bid price and the number of bidding units based on the predicted downward floating rate;

[0068] The quotation module 850 is used to display the prediction results of the bidding probability based on the final similar bidding sections, optimize the prediction column data and count the quotation distribution.

[0069] In particular, an electronic device is also provided, the electronic device includes a processor and a memory, the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the potential competitor analysis method, for example, to execute the above-described Figure 1 The method comprises steps S100 to S500.

[0070] In particular, a computer-readable storage medium is also provided, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement a potential competitor analysis method, for example, to execute the above-described Figure 1 The method comprises steps S100 to S500.

[0071] The processor is used to complete various control logics of the device, which can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a single-chip microcomputer, an ARM (Acorn RISCMachine) or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination of these components. In addition, the processor can also be any traditional processor, microprocessor or state machine. The processor can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP and / or any other such configuration.

[0072] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0073] As an example, non-volatile storage media can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) as external cache memory. By way of illustration and not limitation, RAM can be obtained in many forms such as synchronous RAM (SRAM), dynamic RAM, (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The disclosed memory components or memories of the operating environment described herein are intended to include one or more of these and / or any other suitable types of memory.

[0074] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium, and the computer program can include the processes of the above-mentioned method embodiments when executed. The storage medium can be a memory, a disk, a floppy disk, a flash memory, an optical storage device, etc.

[0075] The above are only some specific implementation methods of the present application, and any other improvements made based on the concept of the present application are deemed to be within the protection scope of the present application.

Claims

1. A potential competitor analysis method, characterized in that: include: Obtain historical bidding section data and pre-process the historical bidding section data; Input the bidding section data to be bid, and screen the pre-processed historical bidding section data based on the discrete features of the bidding section data to be bid, so as to obtain candidate similar bidding sections; Calculate the text similarity between the bidding section and the candidate similar sections and the numerical deviation of the estimated price of the bidding contract, compare the calculated text similarity and numerical deviation with the preset threshold, and re-screen the candidate similar sections to obtain the final similar sections; The winning bid price and the number of bidding units corresponding to the final similar bidding sections are input into the pre-trained logistic regression model for prediction, and a predicted discount rate is generated based on the winning bid price corresponding to the final similar bidding sections, and the predicted winning bid price and the number of bidding units are obtained based on the predicted discount rate; Based on the prediction results of the bidding probability of the final similar bidding sections, the prediction column data is optimized and the quotation distribution is statistically analyzed.

2. The method according to claim 1, characterized in that The obtaining of historical bid section data and preprocessing of the historical bid section data comprises: The contract estimated price and the bid winning price in the historical bidding data are pre-processed to unify the numerical units.

3. The method according to claim 1, characterized in that The bidding section data to be bid is input, and the pre-processed historical bidding section data is screened based on the discrete features of the bidding section data to be bid, and candidate similar bidding sections are obtained, including: The historical bidding section data is screened according to various features of the bidding section data to be bid, wherein the various features include text features, discrete features, time series features and numerical features.

4. The method according to claim 3, characterized in that The screening of the historical bidding section data according to the multiple characteristics of the bidding section data to be bid includes: Through discrete feature screening, the historical bidding section data is preliminarily screened using the discrete features of region and project transaction classification. The historical bidding section data that pass the screening are candidate similar bidding sections.

5. The method according to claim 1, characterized in that The method of calculating the text similarity between the bidding section and the candidate similar sections and the numerical deviation of the estimated price of the bidding contract, comparing the calculated text similarity and numerical deviation with a preset threshold, and re-screening the candidate similar sections to obtain the final similar sections includes: Calculate the text similarity between the bidding section and all candidate similar sections, compare the calculation result with the preset threshold, and use the text similarity higher than the preset threshold as one of the screening items; Calculate the numerical deviation between the estimated contract price of the bidding section and the estimated contract prices of all candidate similar bidding sections, then compare the calculated result with a preset threshold, and use the numerical deviation less than the preset threshold as another screening item; The candidate similar bid sections that meet all the above screening items at the same time are the final similar bid sections.

6. The method according to claim 1, characterized in that The winning bid price and the number of bidding units corresponding to the final similar bidding section are input into the pre-trained logistic regression model for prediction, and a predicted downward floating rate is generated based on the winning bid price corresponding to the final similar bidding section. The predicted winning bid price and the number of bidding units obtained based on the predicted downward floating rate include: Input the winning bid prices and number of bid units corresponding to all final similar bid sections into the pre-trained logistic regression model for prediction; Use the following formula to calculate the predicted downward floating rate of each final similar bid section, count the downward floating rates of all final similar bid sections, and take the average of all downward floating rates as the predicted downward floating rate: (Winning bid price - contract estimated price) / contract estimated price; The predicted winning bid price is obtained by calculating contract estimated price * predicted discount rate = predicted winning bid price.

7. The method according to claim 1, characterized in that The prediction results of the bidding probability based on the final similar bidding sections, optimizing the prediction column data and statistically analyzing the quotation distribution include: The bidding data is associated through the Guid of the final similar bidding sections, and grouped according to the unit name in the discrete feature; According to the bidding times and bidding section times of each bidding unit in the historical bidding data, the bidding probability of each bidding unit is calculated as follows: Sort the bidders in each group by bid probability from high to low, and select the bidders with high bid probability according to the number of bidders predicted previously. If the amount of valid data is insufficient, update the number of bidders to the corresponding number of bidders; For each bidding unit, calculate (quoted price - contract estimated price) / contract estimated price in each bidding section as the discount rate, obtain the average as the discount rate of the corresponding company's quotation, calculate the contract estimated price * (1 + discount rate) as the corresponding company's quotation, and count the quotation distribution of each bidding unit as the final output.

8. A potential competitor analysis system, characterized in that: include: A preprocessing module, used to obtain historical bid section data and preprocess the historical bid section data; A screening module is used to input the bidding section data to be bid, screen the pre-processed historical bidding section data based on the discrete features of the bidding section data to be bid, and obtain candidate similar bidding sections; The prediction module is used to calculate the text similarity between the bidding section and the candidate similar sections and the numerical deviation of the bidding contract estimated price, compare the calculated text similarity and numerical deviation with the preset threshold, and re-screen the candidate similar sections to obtain the final similar sections; The probability module is used to input the winning bid price and the number of bidding units corresponding to the final similar bidding sections into the pre-trained logistic regression model for prediction, generate a predicted downward floating rate based on the winning bid price corresponding to the final similar bidding sections, and obtain the predicted winning bid price and the number of bidding units based on the predicted downward floating rate; The quotation module is used to display the prediction results of bidding probability based on the final similar bidding sections, optimize the prediction column data and count the quotation distribution.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the potential competitor analysis method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are called and executed by the processor, the computer-executable instructions prompt the processor to implement the potential competitor analysis method described in any one of claims 1 to 7.

Images