KR20200028803A - System for calculating interval probability using bidding trend and user behavior pattern - Google Patents

Abstract

A bidding trend analysis device is disclosed. According to an embodiment of the present invention, the bidding trend analysis device comprises: a bidding information extraction unit for extracting bidding information of at least one company from a bidding information database storing existing bidding information; a business bidding assessment rate analysis unit for calculating an assessment rate and statistical information of the assessment rate based on the bidding information; and a bidding value calculating unit for calculating an optimal bidding value based on the calculation result and business bidding assessment rate of the business bidding assessment rate analysis unit.

Description

System for deriving probability of interval by analyzing bidder trend and behavior pattern {SYSTEM FOR CALCULATING INTERVAL PROBABILITY USING BIDDING TREND AND USER BEHAVIOR PATTERN}

According to an embodiment of the present invention, the optimal bidding section is calculated by analyzing bidder tendency and behavior pattern by providing optimal bidding values to the user by analyzing the results of bidding bidding companies' bidding rate and scheduled price bidding rate. It is a related invention.

In general, a government agency or a public agency selects a supplier through a bidding process when it receives more than a certain amount of construction, services, and goods. Electronic bidding has become common in recent years.

In particular, in the electronic procurement system (based on the country market), announcements are posted from the ordering company, and companies suitable for the industry and region participate in the bidding, draw two lottery numbers, and submit bids based on the base amount of the announcement. The lottery number is a standard for determining the multiple reserve price of the ordering party, and in the case of local governments, it is divided evenly from 97% to 103% to the 8th decimal place. The bidding number selected by the companies participating in the bidding is based on the highest number, and the arithmetic average of the base price is the estimated price. The assessment rate is the basic price divided by the expected price.

The current Nara market bidding system follows the minimum bidding principle, that is, a bidding company with a bidding rate that is as low as possible among the ratings higher than the expected price evaluation rate.

The conventional successful bid prediction system collects past information from 15 multiple reserve prices randomly arranged and selected up to 8 decimal places per country for 1,365 multiple reserve price combinations to predict or order random values as the most frequent data, We used a method of predicting the possible rate of assessment by using the multiple reserve prices of the bidding officer and the completed bid. However, this method is meaningless because it is no longer predictable due to the increased difficulty of random numbers due to the recent reorganization of the country market system.

Other successful bid prediction services that do not use multiple reserve price combinations are, in the end, the majority of programs based on the assessment of assessment rates. However, the assessment rate arising from each announcement is an independent event and the individual announcement does not affect the assessment rate of the next announcement.

In addition, individual bidders of small businesses currently have a name, number of companies, base amount, expected price, assessment rate, license, region, preliminary scope, trade name, business, and announcement number-order in the vast public announcement that is almost impossible to be done manually. , Public institutions, input date and time, bid closing date, bid result value extraction and data processing, statistical processing, it takes a lot of time and technical difficulties to generate a large amount of data to statistical analysis, and also focus on construction and services We are experiencing inefficiencies in the operational aspects of the performing companies.

As related prior art, there is Patent Publication No. 2002-0043463 (Invention name: Method for calculating bid amount for electronic bidding and method and system for providing online bidding service using the same, Publication date: 2002.06.10.).

The purpose of an embodiment of the present invention is based on a statistical analysis of randomly generated random rate generated by multiple preliminary prices of an existing bidding company, or based on a statistical analysis of randomly assessed random rate, and not based on the predicted rate. To predict bid propensity and behavior patterns. In addition, after analyzing the ratio of the bidder distribution according to the probability value of the normal distribution section of the random number assessment rate, it predicts and selects a bidding section with a high probability of winning the bid and provides the bidding value to the user.

The bid propensity analysis apparatus according to an embodiment of the present invention for achieving the above object includes: a bid information extraction unit extracting bid information of at least one company from a bid information database storing existing bid information; A business bidding assessment rate analysis unit that calculates a business bidding assessment rate, assessment rate, and respective statistical information based on the bid information; And a bidding value calculating unit that calculates an optimal bidding value based on the calculation result of the firm bidding assessment rate analysis unit.

The bidding information extracting unit extracts bidding information based on at least one of a type of license, the presence or absence of a license, an area, an estimated price, and an ordering system, and the business bidding rate analysis unit and the bidding value calculating unit are licensed , Based on the presence or absence of a license, region, estimated price, and ordering system, results are calculated.

The business evaluation rate and statistical information of the evaluation rate include at least one of each standard deviation, variance, average, and probability distribution.

Preferably, if the number of companies constituting the bidding information is greater than or equal to a predetermined number, the business bidding assessment rate analysis unit may include an example where the business bidding assessment rate, assessment rate and respective statistical information of the bidder of the relevant announcement are randomly divided. A macro-analysis unit that performs a macro-analysis that analyzes the tendency of the bidder distribution distributed over time to change over time; And if the number of companies constituting the bidding information is less than a predetermined number, perform a micro-analysis that analyzes the bidding tendency of each bidder based on the bidding company bidding rating rate, rating rate, and statistical information of each bidder. It includes a microanalysis unit.

The macro-analysis unit inspects the company's investment assessment rate and assessment rate of all public announcements extracted based on at least one of the type of license, the presence or absence of a license, the region, the estimated price, and the ordering system in the bidding information extraction unit Calculate the assessment rate and statistical information of the assessment rate.

The microanalysis unit calculates statistical information of a business bidding assessment rate of the bidding announcement of at least one company extracted based on the type of license, the presence or absence of the license, the estimated price, and the ordering system. The individual company's business evaluation rate stored in chronological order is analyzed by a pattern based on past behavior, and then reclassified and stored.

Preferably, it further includes a verification unit that compares the optimal bid value with a successful bid price.

A method of analyzing bidder propensity and behavior patterns according to an embodiment of the present invention for achieving the above object comprises: extracting bidding information of at least one company from a bidding information database storing existing bidding information; Calculating a business bidding assessment rate, assessment rate and respective statistical information based on the bidding information; And calculating an optimal bidding value based on the firm bidding assessment ratio, the assessment ratio, and each statistical information.

The step of extracting the bidding information includes extracting bidding information based on the type of license, the presence or absence of the license, the region, the estimated price, and the ordering system, and calculating the business bidding assessment rate, assessment rate, and respective statistical information And the step of calculating the optimal bid value calculates the result based on the type of license, the presence or absence of the license, the area, the estimated price, and the ordering system.

The business evaluation rate and statistical information of the evaluation rate include at least one of each standard deviation, variance, average, and probability distribution.

Preferably, the step of calculating the business bidding assessment rate, assessment rate, and each statistical information, if the number of the companies constituting the bidding information is a predetermined number or more, the business bidding assessment rate, assessment rate of each announcement and A macroscopic analysis is performed to analyze the tendency of the bidder distribution distribution, which is distributed according to the preliminary range of segmentation where the statistical information is randomly divided, and if the number of companies constituting the bidding information is less than a predetermined number, the specific company A micro-analysis that analyzes each company's propensity to conduct inspections is performed based on each company's evaluation rate, evaluation rate, and statistical information.

In the step of extracting the bidding information of the at least one company, the macroscopic analysis is based on the type of license, the presence or absence of the license, the region, the estimated price, and the company's investment assessment rate and assessment rate of the announcement extracted based on the ordering system. Statistical information of the ejaculation rate and ejaculation rate is calculated. In addition, we analyze the tendency of the distribution of bidders with randomly divided examples distributed over time to change over time.

The micro-analysis unit statistics of the business bidding assessment rate, assessment rate, and the business bidding assessment rate, assessment rate of at least one bidding announcement extracted based on the type of license, the existence of the license, the region, the estimated price, and the ordering system The information is calculated in chronological order. In addition, the rate of evaluating the company's bidding for individual companies stored in chronological order is analyzed by a pattern based on past behaviors, and then reclassified and stored.

Preferably, the method further includes comparing the optimal bid value with a successful bid price.

The apparatus and method for analyzing a bidding trend according to an embodiment of the present invention extracts the existing bidding information according to at least one item, and randomly divides the extracted bidding information based on the company bidding assessment rate and the statistical information of the assessment rate It is divided into macro analysis that predicts the distribution trend of participating companies in the example price section and micro analysis that analyzes the investment pattern of a specific company, and analyzes the bidding trend and pattern to calculate the highest probability of the bidding section to the user and provides the final review value can do.

1 is a block diagram illustrating a bidding trend and behavior pattern analysis apparatus according to an embodiment of the present invention.
2 is a block diagram illustrating a bidding trend and behavior pattern analysis apparatus according to another embodiment of the present invention.
3 is a block diagram illustrating a bidding trend and behavior pattern analysis apparatus according to another embodiment of the present invention.
4 is a view illustrating a bidding trend and behavior pattern analysis method according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a bidding trend and behavior pattern analysis apparatus according to an embodiment of the present invention.

Referring to FIG. 1, the bidder trend and behavior pattern analysis apparatus 100 according to an embodiment of the present invention includes a bid information extraction unit 110, a company bidding assessment rate analysis unit 120, and a bidding value calculation unit 130. Includes.

The bidding information extraction unit 110 extracts bidding information of at least one company from a bidding information database that stores existing bidding information.

The bidding information database (not shown) is a database included in a server on which electronic bidding is performed, and refers to a storage device that contains information about bidding that has been performed in the previous bidding process. For example, electronic procurement and electronic bidding systems such as Nara Market, Korea Water Resources Corporation and ETRI may be applicable.

The bidding information is necessary information in a series of bidding processes, and includes information such as the number of the bidding announcement, the place of order, the basic amount, the date and time of the bidding, the bidding rate of the company, the date and time of the bidding, and the qualification conditions. Preliminary prices, bidding price and bidding number for each bidder, planned price, successful bid price, lower bid price may be included. In addition, it may include information such as the type of license, the presence or absence of a license, the area, etc., which are information of the bidder, and identification information of the system in which the bidding is performed.

The business bidding assessment rate analysis unit 120 calculates the business bidding assessment rate, assessment rate, and respective statistical information based on the bidding information.

Here, the assessment rate and the assessment rate of the company's bidding may be expressed as [Equation 1] and [Equation 2] below.

[Equation 1]

[Equation 2]

Company bidding assessment rate = Company bidding amount ÷ Basic amount ÷ Lower bidding rate

Here, the statistical information may include standard deviation, variance, average, and probability distribution.

For example, if only the assessment rate information is collected in all the bidding information of a construction that requires a specific license, the procurement agency forms a range of 98% to 102%, and the municipality ranges from 97% to 103%. The probability distribution of the assessment rate can be calculated within the applicable example range.

The bidding value calculating unit 130 calculates an optimal bidding value based on the calculation result of the company bidding assessment rate analysis unit 120.

Here, the optimal bidding value means a bidding value that has a high probability of a successful bid and can generate a high profit. The optimal projection value can be calculated by the following [Equation 3].

[Equation 3]

Optimal bidding value = Basic amount of money × Lower bidding rate × Optimal bidding assessment rate

The optimal bidding assessment rate of [Equation 3] refers to a bidding assessment ratio of a company selected to calculate an optimal bidding value.

If the above expression (3) is expressed differently, various section probabilities may be determined according to the company survey assessment rate, assessment rate and each statistical information calculated by the company survey assessment rate analysis unit 120, and how the section probability is determined Depending on whether or not, the optimal projection section may be calculated differently.

Here, the selection of the optimal survey assessment rate may be selected by a user selection input, or may be selected by an algorithm based on the results of the business survey assessment rate analysis unit 120. There may be various embodiments of an algorithm for determining a business survey assessment rate from the results of the business survey assessment rate analysis unit 120. As one example, a company's bidding rate or predetermined range is selected according to a predetermined criterion or a predetermined algorithm by analyzing the tendency for the company's bidding rate to be distributed in the preliminary range according to each system. The company bidding assessment rate section may be selected, where the selection according to a predetermined algorithm calculates the probability value of the normal distribution of the appearance assessment rate for each divided section, and then analyzes the trend of the bidder as a macro-analysis and the behavior of the bidder as a micro-analysis Pattern analysis is performed. Therefore, a preliminary range may be selected based on a ratio of a bidder's bidding distribution to a probability value of a normal distribution of a randomly divided section, and a bidding value may be determined based on the selected preliminary range.

In addition, the bidding information extracting unit 110 may extract bidding information based on information on the type of license of the bidding company, the presence or absence of the license, and the region and the estimated price and ordering system.

That is, in the bidding information database, a large number of bidding information is stored including various items, and it is inefficient to analyze all of them in bulk. Therefore, it is desirable to selectively extract the type of announcement that the user wants to bid and the essential data from the announcement.

Likewise, when the bidding information extraction unit 110 selectively extracts the bidding information, the company bidding assessment rate analysis unit 120 and the bidding value calculating unit 130 according to the same criteria, the type of license, the presence or absence of the license, the region , The result can be calculated based on the estimated price and ordering system.

2 is a block diagram illustrating a bidder trend and behavior pattern analysis apparatus according to another embodiment of the present invention.

Referring to FIG. 2, the bidder trend and behavior pattern analysis apparatus 100 according to another embodiment of the present invention includes a bid information extraction unit 110, a macro analysis unit 122, a micro analysis unit 124, and a bid value calculation unit It may include (130).

The bidder trend and behavior pattern analysis apparatus 100 according to another embodiment of the present invention, except for the macro-analysis unit 122 and the micro-analysis unit 124, and the bidder trend and behavior pattern analysis apparatus 100 of FIG. same. Therefore, in this embodiment, only the macro-analysis unit 122 and the micro-analysis unit 124 are described in detail, and the description of the remaining components is omitted.

When the number of companies constituting the bidding information to be analyzed is greater than a predetermined number, the macroscopic analysis unit 122 determines the bidding rate of the bidder of the relevant announcement. Perform macro analysis to calculate and analyze the assessment rate and each statistical information.

More specifically, the macro-analysis unit 122 statistics on the assessment rate of the assessment rate of all public announcements extracted based on the type of license, the presence or absence of the license, the region, the estimated price, and the ordering system in the bidding information extraction unit 110 Information can be calculated.

For example, in all announcements that require the same license, it is virtually difficult to analyze the behavioral patterns of each of the numerous companies. Therefore, for example, if the number of companies is N or more, the example of the ordering party, not the behavior pattern for each company, randomly divides the range section and calculates the bidder distribution chart of the bidders in the divided section. In addition, the tendency for the distribution of the distribution in the arbitrarily divided preliminary range to be changed over time is analyzed using statistical information of the company evaluation rate and the company evaluation rate.

When the number of companies constituting the bidding information is less than a predetermined number, the micro-analysis unit 124 calculates the company bidding assessment rate, assessment rate, and statistical information for each specific company.

More specifically, the micro-analysis unit 124 is the order of time of the company bidding assessment rate of the announcement bid by each of at least one company extracted based on the type of license, the presence or absence of the license, the region, the estimated price, and the ordering system. According to the analysis, it is possible to calculate statistical information of the company's inspection assessment rate. In addition, the rate of evaluating the company's bidding for individual companies stored in chronological order is analyzed by a pattern based on past behavior, and then reclassified and stored.

3 is a block diagram illustrating a bidder trend and behavior pattern analysis apparatus according to another embodiment of the present invention.

Referring to FIG. 3, the bidder trend and behavior pattern analysis apparatus 100 according to another embodiment of the present invention includes a bid information extraction unit 110, a company bidding assessment rate analysis unit 120, and a bidding value calculation unit 130 And a verification unit 140.

The bidder trend and behavior pattern analysis apparatus 100 according to another embodiment of the present invention is the same as that of FIG. 1 except for the verification unit 140, so in this embodiment, only the verification unit 140 Explain.

The verification unit 140 compares the optimal bid value with the successful bid price.

Specifically, the verification unit 140 may verify based on the following [Equation 4] whether the optimal bid value suggested by the bid value calculation unit 130 could be a successful bid if the actual bid was actually made in the already announced announcement. .

[Equation 4]

Lower bid price ≤ optimal bid value <successful bid price

The verification unit 140 may consider that the verification has passed only when the optimal projection value satisfies [Equation 4]. However, as another embodiment, if it is within a predetermined error range based on the successful bid price, it may be regarded as passing the verification.

4 is a view illustrating a method for analyzing a bidder trend and behavior pattern according to an embodiment of the present invention.

1 and 4, in step 410, the bidder trend and behavior pattern analysis apparatus 100 extracts bidding information of at least one company from a bidding information database that stores existing bidding information.

Next, in step 420, the bidder trend and behavior pattern analysis apparatus 100 calculates a business bidding assessment rate, assessment rate, and respective statistical information based on the bidding information.

Next, in step 430, the bidder tendency and behavior pattern analysis apparatus 100 calculates an optimal bidding value based on the company bidding assessment rate, assessment rate, and each statistical information.

So far, the present invention has been focused on the preferred embodiments. Those skilled in the art to which the present invention pertains will understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered in terms of explanation, not limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the equivalent range should be interpreted as being included in the present invention.

Claims (14)

A bidding information extraction unit for extracting bidding information of at least one company from a bidding information database storing existing bidding information;
A business bidding assessment rate analysis unit that calculates a business bidding assessment rate, assessment rate, and respective statistical information based on the bid information; And
Bidder trend and behavior pattern analysis apparatus including a bid value calculation unit for calculating an optimal bid value based on the calculation result of the business bidding assessment rate analysis unit. According to claim 1,
The bidding information extraction unit
Bid information is extracted based on the type of license, the presence or absence of the license, the region, the estimated price, and the ordering system.
The company bidding rate analysis unit and the bidding value calculation unit
A bidder trend and behavior pattern analysis device that calculates results based on the type of license, the presence or absence of a license, the area, the estimated price, and the ordering system. According to claim 1,
The bidding tendency and behavior pattern analysis device, wherein the company bidding assessment rate and statistical information of the assessment rate include at least one of each standard deviation, variance, average, and probability distribution. According to claim 1,
The company's bidding assessment rate analysis department
When the number of companies constituting the bidding information is greater than or equal to a predetermined number, the bidding rate of the relevant announcement is calculated according to the business bidding rate, the rating rate, and each statistical information, and the statistical information is distributed according to a randomly divided example range A macro-analysis unit that performs macro-analysis to analyze the tendency of the bidder distribution to change over time; And
When the number of companies constituting the bidding information is less than a predetermined number, the bidding company bidding rate, assessment rate and respective statistical information for each specific company are calculated, and each company's bidding tendency is analyzed based on the statistical information. Bidder trend and behavior pattern analysis apparatus including a micro-analysis unit for performing a micro-analysis. According to claim 4,
The macro analysis unit
The bidding information extraction unit based on the type of license, the presence or absence of a license, the region, the estimated price, and the bidding company's bidding assessment rate and assessment rate of the relevant announcement extracted based on the ordering system Bidder trend and behavior pattern analysis device for calculating statistical information. According to claim 4,
The microanalysis unit
Statistical information on the bidder's bidding rate, assessment rate, and bidder's bidding assessment rate, assessment rate, based on the type of license, the presence or absence of the license, the estimated price, and the ordering system extracted from at least one company Bidder trend and behavior pattern analysis device that calculates and stores the data in order of time and reclassifies them according to each pattern. According to claim 1,
A bidder trend and behavior pattern analysis apparatus further comprising a verification unit comparing the optimal bid value with a successful bid price. Extracting bidding information of at least one company from a bidding information database storing existing bidding information;
Calculating a business bidding assessment rate, assessment rate and respective statistical information based on the bidding information; And
A method of analyzing bidder trends and behavior patterns, including calculating an optimal bid value based on a business bidding assessment rate, a rating ratio, and each statistical information and a captured business bidding assessment section. The method of claim 8,
The step of extracting the bid information
Bid information is extracted based on the type of license, the presence or absence of the license, the region, the estimated price, and the ordering system.
The step of calculating the company bidding assessment rate, the assessment rate and the respective statistical information, and the step of calculating the optimal bidding value are
A method of analyzing bidding trends and behavior patterns to calculate results based on the type of license, the presence or absence of a license, the area, the estimated price, and the ordering system. The method of claim 8,
A method of analyzing bidder trends and behavior patterns, wherein the business evaluation rate and statistical information of the evaluation rate include at least one of each standard deviation, variance, average, and probability distribution. The method of claim 8,
The step of calculating the assessment rate, assessment rate, and statistical information of each of the companies
When the number of companies constituting the bidding information is greater than or equal to a predetermined number, the bidders of the relevant announcement calculate bidding rate, assessment rate, and statistical information of each bidder, and the bidders distributed according to the preliminary range in which the statistical information is arbitrarily divided A macroscopic analysis is performed to analyze the tendency of the distribution to change over time.
If the number of companies constituting the bidding information is less than a predetermined number, the bidding company bidding rate, assessment rate, and statistical information for each specific company are calculated, and the bidding propensity of each company is calculated based on the statistical information. A method of analyzing bidder trends and behavior patterns to perform micro-analysis. The method of claim 11,
The macroscopic analysis is
In the step of extracting the bidding information of the at least one company, the company is assessed by the type of license, the presence or absence of the license, the region, the estimated price, and the business review rate and assessment rate of the relevant advertisement extracted based on the ordering system A method of analyzing bidders' tendencies and behavior patterns that statistically analyzes the propensity to change the distribution of bidding rate assessment rate over time by calculating statistical information of rate and assessment rate, and randomly divided examples are distributed over a range section. The method of claim 11,
The microanalysis unit
Based on the type of license, the presence or absence of the license, the region, the estimated price, and the ordering system, statistics information of the company's bidding rate, assessment rate, and the company's bidding assessment rate and assessment rate of at least one bidding announcement A bidder trend and behavior pattern analysis method that calculates in accordance with the order of the bidders to grasp the tendency of the bidding pattern according to a certain formula based on the past bidding rate data of each bidder. The method of claim 8,
A method of analyzing bidder trends and behavior patterns, further comprising comparing the optimal bid price with a successful bid price.

Images