KR20220083183A - System for supporting bidding strategy establishment using personalization based business recommendation algorithm and method thereof - Google Patents

Abstract

The present invention relates to a bidding strategy establishment support system and method using a personalization-based business recommendation algorithm, and more particularly, to a database by collecting big data including not only structured data but also unstructured data from a plurality of external servers. And, by using a business recommendation algorithm according to the individual characteristics of the user to strategically recommend a bid announcement by correcting the weight for each analysis element through learning of the hit rate and hit rate trend for each analysis element by interaction with the user, the user A bidding strategy establishment support system and method using a personalization-based business recommendation algorithm capable of recommending a customized business or a business with a high possibility of winning orders are provided.

Description

Bid strategy establishment support system and method using personalization-based business recommendation algorithm

The present invention relates to a bidding strategy establishment support system and method using a personalization-based business recommendation algorithm, and more particularly, to a database by collecting big data including not only structured data but also unstructured data from a plurality of external servers. And, by using a business recommendation algorithm according to the individual characteristics of the user to strategically recommend a bid announcement by correcting the weight for each analysis element through learning of the hit rate and hit rate trend for each analysis element by interaction with the user, the user It relates to a bidding strategy establishment support system and method using a personalization-based business recommendation algorithm capable of recommending a customized business or a business with a high probability of winning.

With the advent of a hyper-connected society that connects everything, R&D actors in each field, such as countries and companies, want to develop big data technology as a future strategic technology to utilize all data generated in the process of connection and contact as valuable. Trying hard. In particular, global companies are using big data not only to solve various problems and respond to issues, but also as an important tool for future strategies and accompanying strategic decisions. According to 2015 WIKIBON data, the global big data market is expected to exceed 84.6 billion dollars in 2026, and the domestic big data market is 262.3 billion won as of 2015, and 30.2% of companies wanting to introduce big data is being investigated as Currently, numerous big data related software is being researched and developed according to the growth potential and demand of big data.

In particular, in the field of information communication, that is, wired and wireless communication technologies and various mobile devices are rapidly developing, it is based on the development of data accumulation analysis and utilization methods. Big data technology, which collects and analyzes data to predict the future and create new values, is now going beyond the introduction stage and is gradually expanding. Big data technology is being widely applied to strengthen the competitiveness of existing industries and the efficiency of public services, and to create and innovate new service products.

Meanwhile, public institutions such as the Public Procurement Service are making efforts to develop public systems by reflecting big data technology in the electronic procurement ordering system, etc. I have been trying to do the back. However, no clear results have been achieved so far.

In addition, various public institutions that carry out procurement are operating an electronic procurement system to support ordering by demanding organizations and procurement companies. The electronic procurement system is a service that systematizes all tasks related to public project order support and supports it through an information system. It is difficult to easily find procurement announcements that are rapidly increasing.

In particular, users of the procurement notice system have been complained of inconvenience because it takes a lot of time and effort to read and review the contents one by one in order to find a procurement notice suitable for bidding in the company to which one belongs. In order to solve this problem, a system and service that can recommend an appropriate procurement notice by analyzing the contents of the procurement notice or the user's company information are required.

Korean Patent Registration [10-1991178] discloses a procurement notification recommendation method and system using collaborative filtering.

Korean Patent Registration [10-1991178] (Registration Date: 2019. 06. 13)

Accordingly, the present invention has been devised to solve the above problems, and an object of the present invention is to collect big data including not only structured data but also unstructured data from a plurality of external servers to build a database, and By using the business recommendation algorithm according to the individual characteristics of the user, and by strategically recommending a bid notice by correcting the weight of each analysis element through learning of the hit rate and hit rate trend for each analysis element by interaction with the user, a business tailored to the user Alternatively, to provide a bidding strategy establishment support system and method using a personalization-based business recommendation algorithm capable of recommending a business with a high possibility of winning an order.

The purpose of the embodiments of the present invention is not limited to the above-mentioned purpose, and other objects not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description. .

In order to achieve the above object, the bidding strategy establishment support system using a personalization-based business recommendation algorithm according to an embodiment of the present invention uses a business recommendation algorithm according to the individual characteristics of the user to interact with the user. a bidding strategy collection support server 100 for correcting the weight for each analysis element through learning of the hit rate and hit rate trend for each analysis element by the analysis element, recommending a corresponding bid announcement, and establishing a customized bidding strategy; a plurality of user terminals (110, 120) for accessing the bidding strategy collection support server, transmitting and receiving data necessary to receive a customized business announcement, and receiving a customized business announcement according to the business recommendation algorithm; and a plurality of external data servers 130 and 140 for collecting data necessary to establish and provide a bidding strategy according to the customized business announcement, wherein the bidding strategy collection support server 100 includes the plurality of user terminals an interface unit 101 that provides an interface for providing the customized business announcement; a data collection unit 103 for collecting and processing data from the plurality of external data servers, extracting keywords, inputting them into a database, and extracting individual characteristic elements from an input signal through the user terminal; a data processing unit 104 for performing data processing necessary to provide the customized business announcement based on data transmitted and received through the interface unit and data collected through the data collection unit; an artificial intelligence analysis unit 105 for updating the weight model by analyzing the hit ratio and the hit ratio trend according to the selection input of the user terminal and the bidding result; a database management unit 106 that stores data necessary to provide the customized business announcement; and a control unit 102 for controlling each component including the interface unit, data collection unit, data processing unit, artificial intelligence analysis unit, and database management unit.

The data collection unit 103 includes: a collection unit 301 for collecting structured data and unstructured data from the plurality of external data servers; a processing unit 302 for processing the collected data; a keyword extraction unit 303 for extracting keywords from the processed data; a DB input unit 304 for inputting the extracted keyword into a database; a data conversion unit 305 for receiving and converting the input signal of the user terminal; and a user characteristic extracting unit 306 for extracting a user characteristic element from the converted user terminal input signal.

The data processing unit 104 includes: a business size determination unit 401 for determining the business size of the business announcement to be recommended to the user from the data stored in the database; a business type determination unit 402 for determining a business type of a business announcement to be recommended to the user from the data stored in the database; a detailed business type determination unit (403) for determining a detailed business type of a business announcement to be recommended to the user from the data stored in the database; a customer type determining unit 404 for determining a customer type of a business advertisement to be recommended to the user from the data stored in the database; a business location determination unit (405) for determining a business location of a business announcement to be recommended to the user from the data stored in the database; ?? A project for determining the business recommendation scope and criteria based on the determined business size, business type, detailed business type, customer type and business location, and calculating an evaluation value for each element of the target business included in the business recommendation scope and criteria It is characterized in that it includes an evaluation value calculation unit 406 for each element.

The artificial intelligence analysis unit 105 may include: a business announcement recommendation unit 501 for recommending a business announcement according to a total score summed using an evaluation value for each element of the business for the target business and a weight for each element; a bidding result analysis unit 502 for analyzing the bidding result based on the user's satisfaction feedback for the recommended project announcement and the following bidding result database; a hit ratio calculation unit 503 for calculating a hit ratio for each analysis element from the bidding result; a learning unit 504 for learning the characteristics of the user's (company) order receiving business based on the calculated hit ratio and the hit ratio trend according to the past history; an inference unit 505 for extracting and inferring factors related to successful bids from the learned characteristics of the order receiving business; a weight modeling unit 506 for modeling a weighting method for each analysis element according to the inference result; and a weight adjustment unit 507 for each element for adjusting a weight for each analysis element.

The project size determination unit 401 is characterized in that the project size is determined in a preset amount range by calculating the distribution, average value, and trend of each type of existing project, and the project type determination unit 402 is , characterized in that the proportion (ratio) of the business type is determined by preference according to the business type ratio of "service: article: construction" of the previously performed business, and the detailed business type determining unit 403 is It is characterized in that the detailed business type is determined through the keyword analysis of the announcement name of the business, and the customer type determining unit 404 determines the proportion (ratio) of the customer type with preference according to the customer type of the existing business. Characterized in that, the business location determining unit 405 determines using the location (location) of the user (company) and the business location of the customer (ordering organization), and the evaluation by business element The value calculating unit 406 divides a section by a preset amount unit based on the past history of the user (company) with respect to the business scale, counts the number of projects for each section, and 70% of the total number of projects received in the past Detect the project cost section in which is distributed, call it P1, calculate “average value of P1 = sum of project expenses in P1 / number of projects in P1”, and determine if the absolute value of the difference between the project cost of the recommended project and the average value of P1 is close to 0. It is characterized in that the evaluation value is calculated by giving the highest point recorded, and with respect to the above business type, the weight of recommendation for each project type is derived by adjusting the deviation of the derived weight of each type (product: construction: service), For the detailed business type, the keywords of the corresponding company are analyzed, the duplicate keyword sequence detection and score of the first keyword of the existing business order are detected and the duplicate keyword sequence of the second keyword searched by the user (company) for the last 1 month Detect and assign a score, assign weights to the duplicated keywords of the second keyword, assign a final keyword ranking and score of the company, extract and analyze keywords for each recommended business, assign a score to the keyword, , luck It is characterized in that the evaluation value is calculated by adding additional points to the number of keywords, and with respect to the customer type, the weight of the ordering organization type at the time of business recommendation is derived by adjusting the deviation of the determined percentage (ratio) of the customer type It is characterized in that, with respect to the location of the business, it is characterized in that the distance index that is the absolute value of the difference between the location code of the company and the location code of the business is calculated.

In addition, the method for supporting the establishment of a bid strategy using a personalization-based business recommendation algorithm in the form of a program executed by a calculation processing means including a computer according to an embodiment of the present invention for achieving the object as described above includes a plurality of A data collection step (S710) of collecting data necessary for establishing a bidding strategy from an external data server; A data processing step (S720) of acquiring and processing the user's past business history information from the collected data to determine the business recommendation range and criteria including the priority and weight of each element to which the user's individual characteristics are applied (S720); a factor-by-element evaluation value calculation step (S730) of calculating an evaluation value for each element of the target project included in the project recommendation range and standard; a business announcement recommendation step (S740) of recommending a business announcement according to a total score summed using the evaluation value for each element of the business and the weight for each element for the target business; element-specific characteristic analysis step (S750) of analyzing element-specific characteristics according to the user's satisfaction feedback and bidding results for the recommended project announcement; a weight correction step (S760) of correcting the weight for each element by learning the hit ratio for each element and the transition learning of the hit ratio for each element; and a bidding strategy providing step (S770) of providing a bidding strategy by repeating the project recommendation step, the element-specific characteristic analysis step, and the weight correction step.

The data collecting step (S710) includes: a collecting step (S810) of collecting structured data and unstructured data from the plurality of external data servers; A processing step of processing the collected data (S820); A keyword extraction step of extracting a keyword from the processed data (S830); DB input step of inputting the extracted keyword into the database (S840); a data conversion step of receiving and converting the input signal of the user terminal (S850); and a characteristic element extraction step (S860) of extracting characteristic elements from the converted input signal of the user terminal.

The data processing step (S720), the business scale determining step (S910) of determining the business scale of the business announcement to be recommended to the user; a business type determination step (S920) of determining the business type of the business announcement to be recommended to the user; a detailed business type determination step of determining the detailed business type of the business announcement to be recommended to the user (S930); a customer type determination step (S940) of determining a customer type of a business advertisement to be recommended to the user; a business location determination step (S950) of determining a business location of a business announcement to be recommended to the user; and a recommendation range determination step (S960) of determining a business recommendation scope and criteria based on the determined business scale, business type, detailed business type, customer type, and business location.

The project size determination step (S910) is characterized in that the project size is determined in a preset amount range by calculating the distribution, average value, and trend of each type of the existing project, and the project type determination step (S920) is , characterized in that the proportion (ratio) of the business type is determined by preference according to the business type ratio of "service: article: construction" of the existing project, and the detailed business type determination step (S930) is, It is characterized in that the detailed business type is determined through the keyword analysis of the announcement name of the business, and in the customer type determination step (S940), the proportion (ratio) of the customer type is determined by preference according to the customer type of the existing business characterized in that, the business location determination step (S950), characterized in that the determination using the location (location) of the user (company) and the business location of the customer (ordering organization), the evaluation value for each element In the calculation step (S730), with respect to the business scale, sections are divided by a preset amount unit based on the past history of the user (company), and the number of projects for each section is counted, so that 70% of the total number of projects received in the past Detect the distributed project cost section and call it P1, calculate “average value of P1 = sum of project cost in P1 / number of projects in P1” It is characterized in that the evaluation value is calculated by giving the highest point, and with respect to the project type, the weight of recommendation for each project type is derived by adjusting the deviation of the derived weight of each type (product: construction: service), For detailed business types, the keywords of the company are analyzed, the duplicate keyword sequence detection and score of the first keywords of the existing project are awarded, and the duplicate keyword sequence of the second keyword searched by the user (company) in the last 1 month is detected and assigning a score, assigning weights to the duplicated keywords of the second keyword, assigning a final keyword ranking and score of the company, extracting and analyzing keywords for each recommended business, and assigning scores to the corresponding keywords; It is characterized in that an additional point is applied to a plurality of keywords to calculate an evaluation value, and with respect to the customer type, the weight of the ordering organization type at the time of business recommendation is derived by adjusting the deviation of the determined percentage (ratio) of the customer type It is characterized in that, with respect to the location of the business, it is characterized in that the distance index that is the absolute value of the difference between the location code of the company and the location code of the business is calculated.

The weight correction step (S760) includes: a hit ratio calculation step (S1010) of calculating a hit ratio for each element for each analysis element from the bidding result; a learning step (S1020) of learning the characteristics of the user (company)'s order receiving business based on the calculated hit ratio and the hit ratio trend according to the past history; an inference step (S1030) of extracting and inferring factors related to successful bids from the learned characteristics of the order receiving business; a weight modeling step of modeling a weighting method for each analysis element according to the inference result (S1040); and an adjustment step (S1050) of adjusting a weight for each of the analysis elements.

In addition, according to an embodiment of the present invention, a computer-readable recording medium storing a program for implementing the method for supporting the establishment of a bidding strategy using the personalization-based business recommendation algorithm is provided.

In addition, according to an embodiment of the present invention, in order to implement the method for supporting the establishment of a bidding strategy using the personalization-based business recommendation algorithm, a program stored in a computer-readable recording medium is provided.

According to the bidding strategy establishment support system and method using the personalization-based business recommendation algorithm according to an embodiment of the present invention, big data including not only structured data but also unstructured data is collected from a plurality of external servers to build a database And, by using a business recommendation algorithm according to the individual characteristics of the user to strategically recommend a bid announcement by correcting the weight for each analysis element through learning of the hit rate and hit rate trend for each analysis element by interaction with the user, the user It has the effect of being able to recommend a customized business or a business with a high probability of winning.

In addition, according to the bidding strategy establishment support system and method using a personalization-based business recommendation algorithm according to an embodiment of the present invention, users do not have to search for business announcements that match their own companies individually. It has the effect of automatically recommending the optimal procurement notice.

1 is a configuration diagram of an embodiment of a bidding strategy establishment support system using a personalization-based business recommendation algorithm according to the present invention.
Figure 2 is a configuration diagram of the bidding strategy establishment support server of Figure 1;
3 is a detailed configuration diagram of the data collection unit of FIG. 2 .
4 is a detailed configuration diagram of the data processing unit of FIG. 2 ;
5 is a detailed configuration diagram of the artificial intelligence analysis unit of FIG. 2 .
6 is a view for explaining the customer types classified in the bidding strategy establishment support system using the personalization-based business recommendation algorithm according to the present invention.
7 is a flowchart illustrating a method for supporting establishment of a bidding strategy using a personalization-based business recommendation algorithm according to the present invention.
8 is a detailed flowchart of an embodiment of the data collection step (S710) of FIG.
9 is a detailed flowchart of an embodiment of the data processing step (S720) of FIG.
10 is a detailed flowchart of an embodiment of the weight correction step (S760) of FIG.

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

When an element is referred to as being “connected” or “connected” to another element, it is understood that it may be directly connected or connected to the other element, but other elements may exist in between. it should be

On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle.

The terms used herein are used only to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, terms such as "comprise" or "have" are intended to designate that a feature, number, process, operation, component, part, or a combination thereof described in the specification exists, but one or more other features It is to be understood that this does not preclude the existence or addition of numbers, processes, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings. Prior to this, the terms or words used in the present specification and claims should not be construed as being limited to conventional or dictionary meanings, and the inventor should properly understand the concept of the term in order to best describe his invention. Based on the principle that can be defined, it should be interpreted as meaning and concept consistent with the technical idea of the present invention. In addition, if there is no other definition in the technical terms and scientific terms used, it has the meaning commonly understood by those of ordinary skill in the art to which this invention belongs, and the summary of the present invention in the following description and accompanying drawings Descriptions of known functions and configurations that may be unnecessarily obscure will be omitted. The drawings introduced below are provided as examples so that the spirit of the present invention can be sufficiently conveyed to those skilled in the art. Accordingly, the present invention is not limited to the drawings presented below and may be embodied in other forms. Also, like reference numerals refer to like elements throughout. It should be noted that the same components in the drawings are denoted by the same reference numerals wherever possible.

An artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike the existing rule-based smart system, the machine learns, judges, and becomes smarter by itself. As artificial intelligence systems are used, the recognition rate improves and users can understand user preferences more accurately.

Artificial intelligence technology consists of machine learning (deep learning) and element technologies using machine learning.

Machine learning is an algorithm technology that categorizes/learns the characteristics of input data by itself, and element technology uses machine learning algorithms such as deep learning to simulate functions such as cognition and judgment of the human brain. It consists of technical fields such as understanding, reasoning/prediction, knowledge expression, and motion control.

The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology for recognizing and applying/processing human language/text, and includes natural language processing, machine translation, dialogue system, question and answer, and speech recognition/synthesis. Visual understanding is a technology for recognizing and processing objects like human vision, and includes object recognition, object tracking, image search, human recognition, scene understanding, spatial understanding, image improvement, and the like. Inferential prediction is a technology for logically reasoning and predicting by judging information, and includes knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Knowledge expression is a technology that automatically processes human experience information into knowledge data, and includes knowledge construction (data generation/classification) and knowledge management (data utilization). Motion control is a technology for controlling autonomous driving of a vehicle and movement of a robot, and includes motion control (navigation, collision, driving), manipulation control (action control), and the like.

Meanwhile, data collection and processing used in the present invention is performed using robotic processing automation (RPA) technology. Robotic Process Automation (RPA) technology is a technology that formalizes and logically automatically performs data generated in the business process. It automates repetitive simple tasks and performs them quickly and precisely.

In other words, it is possible to shorten the overall business time and reduce costs.

Business automation technology using intelligent robot software has been used since the 1990s for screen scraping to collect necessary information from website screens or workflow automation to automatically process and manage all business-related matters.

While general business automation is built as a business management-oriented process that applies artificial intelligence (AI) and machine learning technology, RPA is designed as a rule-based business process from the end user's point of view, so that simple repetitive tasks are constantly performed instead of people. carried out in bulk RPA can only handle tasks for which the current rules are clearly defined, and is not a substitute for human judgment.

By applying cognitive technologies such as machine learning, speech recognition, and natural language processing, it can be used for cancer diagnosis in the medical field that requires human cognitive ability, customer asset management in the financial industry, and analysis of legal precedents.

Big data refers to data sets that go beyond the collection, storage, management, and analysis capabilities of existing data. Big data may be classified into structured data, semi-structured data, and unstructured data according to the degree of standardization.

Structured data refers to data stored in a fixed field. In other words, it refers to data stored in a certain format. Semi-structured data refers to data that is not stored in a fixed field, but includes metadata or schema. Examples of semi-structured data include Extensible Mark-up Language (XML) and Hypertext Mark-up Language (HTML). Unstructured data refers to data that is not stored in a fixed field. Examples of the unstructured data include text documents, image data, video data, and audio data.

The bidding strategy establishment support system according to the present invention may collect and analyze the big data as described above. Examples of big data analysis techniques include text mining, reputation mining, social network analysis, cluster analysis, neural network analysis, and markov model. However, it is not limited to the illustrated analysis techniques.

Text mining is a technology for extracting and processing useful information from semi-structured text data or unstructured text data based on natural language processing technology. Natural language processing (NLP) technology is a technology that enables natural language understanding and natural language generation. Natural language is a term used by humans to communicate, as opposed to artificial language (computer language). Natural language understanding refers to the mechanical analysis of natural language into a form that can be understood by a computer. Natural language generation refers to enabling computers to output natural language. Natural language processing technology consists of morphological analysis, syntax analysis, semantic analysis, and pragmatic analysis.

Reputation analysis is performed on structured text and unstructured text in social media such as blog, social network service (SNS), wiki, handcrafted product (UCC), micro-blog, etc. It is a technology that collects and analyzes text to determine the reputation (eg, positive, negative, neutral) of a product or service.

Social network analysis is a technique for analyzing and extracting a user's influence, interest, and tendency based on a social network connection structure and connection strength. The types of information that can be utilized when performing social network analysis can be broadly classified into follow, reply, and share (retweet, in the case of Twitter). When you want to see his writing, you follow him, and if you comment or share (retweet), it means an active mutual relationship. By analyzing these relationships, you can evaluate the user's influence or dissemination, and furthermore, the spread of specific articles. You can track the process or estimate the spread of future opinions.

Cluster analysis is a technique for finally discovering groups with similar characteristics while merging individuals with similar characteristics.

1 is a configuration diagram of a bidding strategy establishment support system using a personalization-based business recommendation algorithm according to an embodiment of the present invention, and FIG. 2 is a configuration diagram of a bidding strategy establishment support server of FIG. 1 .

Referring to FIG. 1 , a bidding strategy establishment support system using a personalization-based business recommendation algorithm according to an embodiment of the present invention includes a bidding strategy collection support server 100 , a plurality of user terminals 110 and 120 , and an external data server. (130, 140).

The bidding strategy collection support server 100 is a computing system having software and/or hardware components for providing a web server-based or dedicated application-based bidding strategy establishment service, and a plurality of user terminals ( 110 and 120) and external data servers 130 and 140 (eg, shopping mall server, portal server) 150 to provide a bidding strategy to the plurality of user terminals 110 and 120 .

The bidding strategy collection support server 100 uses a business recommendation algorithm according to the individual characteristics of the user to correct the weight for each analysis element through the learning of the hit rate and hit rate trend for each analysis element by interaction with the user. A bid notice is recommended and a customized bid strategy is established and provided to the user terminals 110 and 120 .

The user terminals 110 and 120 access the bidding strategy collection support server 100, transmit and receive data necessary to receive a customized business announcement, and receive a customized business announcement according to the business recommendation algorithm.

The plurality of external data servers 130 and 140 collect data necessary to establish and provide a bidding strategy according to the customized business announcement.

Meanwhile, the bidding strategy establishment support system using the personalization-based business recommendation algorithm according to an embodiment of the present invention may further include a payment server (not shown) for handling the billing of the user terminals 110 and 120 . In addition, the payment server may include a separate database.

Referring to FIG. 2 , the bidding strategy collection support server 100 includes an interface unit 101 , a control unit 102 , a data collection unit 103 , a data processing unit 104 , an artificial intelligence analysis unit 105 , and a database. management unit 106 and the like. The interface unit 101, the control unit 102, the data collection unit 103, the data processing unit 104, the artificial intelligence analysis unit 105, and the database management unit 106 is at least a part of them It may be program modules that communicate with the bidding strategy collection support server 100 . These program modules may be included in the bidding strategy collection support server 100 in the form of an operating system, an application program module, and other program modules, and may be physically stored in various known storage devices. In addition, these program modules may be stored in a remote storage device capable of communicating with the bidding strategy collection support server 100 . Meanwhile, these program modules include, but are not limited to, routines, subroutines, programs, objects, components, data structures, etc. that perform specific tasks or execute specific abstract data types according to the present invention.

Here, the communication network may be configured regardless of its communication mode, such as wired and wireless, and includes a local area network (LAN), a metropolitan area network (MAN), and a wide area network (WAN). It may be composed of various communication networks such as Preferably, the communication network referred to in the present invention may be a well-known World Wide Web (WWW; World Wide Web).

The interface unit 101 interfaces so that the bidding strategy collection support server 100 can communicate with the plurality of user terminals 110 and 120, and graphics necessary for transmitting and receiving data related to the bidding strategy provision. A user interface may be provided to the plurality of user terminals 110 and 120 .

The control unit 102 controls the flow of data between the interface unit 101 as described above, a data collection unit 103 to be described later, a data processing unit 104 , an artificial intelligence analysis unit 105 and a database management unit 106 . perform the control function.

The data collection unit 103 collects and processes data from the plurality of external data servers 130 and 140 , extracts keywords and inputs them to a database, and individual characteristics from input signals through the user terminals 110 and 120 . Extract the element.

The data processing unit 104 performs data processing necessary to provide the customized business announcement based on data transmitted and received through the interface unit 101 and data collected through the data collection unit 103 .

The artificial intelligence analysis unit 105 updates the weight model by analyzing the hit ratio and the hit ratio trend according to the selection input and the bidding result of the user terminals 110 and 120 .

The database management unit 106 stores a member (company) database 106a for managing personal information of a user (company) to receive a customized business announcement as a member, and a past history of the user (member) a past history database 106b for managing and managing, a business announcement database 106c that stores business announcements collected through the data collection unit 103, and a characteristic that extracts and stores user characteristics from signals received from users It may include an extraction database 106d, a bid result database 106e that stores the bidding results for the member's bid, and a weight model database 106f that stores a weight model for recommending the business announcement. .

For example, the member database 106a may include identification information such as login ID and password used by members to log in, and information about name, gender, business number, address, contact information, and the like.

Although not shown in the drawing, the billing database may store billing data according to members. Meanwhile, it may further include billing and payment details according to the membership level.

In the above embodiment, the database for storing information for the implementation of the present invention is a member database 106a, a past history database 106b, a business announcement database 106c, a feature extraction database 106d, and a bidding result database 106e. , and six databases of the weight model database 106f, but the configuration of the database including this classification may be changed according to the needs of those skilled in the art.

Meanwhile, in the present invention, a database is a concept that includes not only a narrow database but also a database in a broad sense including data records based on a computer file system. It should be understood that if the data of the data can be extracted, it can be included in the database according to the present invention.

The plurality of user terminals 110 and 120, on the other hand, the plurality of user terminals 110 and 120 to the bidding strategy collection support server 100 through a communication network of a person interested in a personalization-based customized business notification providing service A digital device including a function for enabling communication after being connected, and having a memory means such as a personal computer (eg, a desktop computer, a notebook computer, etc.), a workstation, a PDA, a web pad, a mobile phone, and the like, and a microprocessor Any number of digital devices equipped with arithmetic capability may be employed as a plurality of user terminals 110 and 120 according to the present invention.

On the other hand, the plurality of user terminals 110 and 120 may be installed with a dedicated application for using a personalization-based customized business notification providing service.

The plurality of user terminals 110 and 120 sign up for membership in order to receive a personalization-based customized business announcement service, request payment of a service fee according to the membership level to a payment server (not shown), and support the collection of the bidding strategy The server 100 requests a personalization-based customized business announcement service, and displays the personalized-based customized business announcement received from the bidding strategy collection support server 100 .

3 is a detailed configuration diagram of the data collection unit of FIG. 2 .

Referring to FIG. 3 , the data collection unit 103 includes a collection unit 301 , a processing unit 302 , a keyword extraction unit 303 , a DB input unit 304 , a data conversion unit 305 , and user characteristic extraction. part 306 .

The collection unit 301 collects structured data and unstructured data from the plurality of external data servers.

The processing unit 302 processes the collected data.

The keyword extraction unit 303 extracts a keyword from the processed data.

The DB input unit 304 inputs the extracted keyword into the database.

The data conversion unit 305 receives and converts the input signal of the user terminals 110 and 120 .

The user characteristic extracting unit 306 extracts a user characteristic element from the converted user input signal of the user terminals 110 and 120 .

4 is a detailed configuration diagram of the data processing unit of FIG. 2 .

Referring to FIG. 4 , the data processing unit 104 includes a business size determination unit 401 , a business type determination unit 402 , a detailed business type determination unit 403 , a customer type determination unit 404 , and a business location determination unit 405 , and an evaluation value calculation unit 406 for each business element.

The business scale determining unit 401 determines the business scale of the business announcement to be recommended to the user from data stored in the database.

The business type determining unit 402 determines the business type of the business announcement to be recommended to the user from the data stored in the database.

The detailed business type determining unit 403 determines the detailed business type of the business announcement to be recommended to the user from the data stored in the database.

The customer type determining unit 404 determines the customer type of the business announcement to be recommended to the user from the data stored in the database.

The business location determining unit 405 determines the business location of the business announcement to be recommended to the user from the data stored in the database.

The evaluation value calculation unit 406 for each business element determines a business recommendation scope and criteria based on the determined business scale, business type, detailed business type, customer type and business location, and is included in the business recommendation scope and criteria Calculate the evaluation value for each element of the target project.

Hereinafter, the determination method for each element will be described in detail.

<Business scale (public notice project cost)>

The business scale is determined by using the project cost for each type of the existing order received from the user (company)'s past history data and the project cost of the business that the user (company) has inquired in detail (link moves).

The project size determination unit 401 analyzes the project cost by project type of the existing order-based project with respect to the maximum project cost for each type, the minimum project cost for each type, the distribution map and section of the project cost for each type, and the trend of the average project cost for each type by year/month.

In addition, the business size determining unit 401 is a distribution map and section of the business cost by type of business that the user (company) has searched-linked in the last 3 months, and the business cost distribution diagram that the user (company) has searched-linked in the last 1 month Analyzes the business cost of the business that the user (company) inquires (moves the link) in detail for the section and the like.

The evaluation value calculation unit 406 for each project element derives a project recommendation range by adjusting the upper limit for each type, the lower limit for each type, the upper/lower limit of the first priority section of distribution, and the recommended section according to the average movement trend.

In addition, the evaluation value calculation unit 406 for each business element is the average value of the 1st ranking section of the search-link moving business criteria distribution for the last 3 months, and the 1st ranking of the business criteria distribution ratio that the user (company) has searched-linked in the last 1 month The project recommendation range is reduced by deriving the average value of the section.

Here, if the average movement trend is on an upward trend, the closer to the average value in the project range equal to or higher than the average value of the 1st priority section, the higher the ranking. The closer to , the higher the ranking, and the evaluation value for the business size is given. At this time, the evaluation value is calculated so that the 1st priority is the highest point and the ranking becomes lower as it goes down.

Here, the evaluation value for the business size can be calculated as follows.

1. Classification of sections by a preset amount based on the company’s past history

2. Count the number of projects for each section

3. The project cost section in which 70% of the total number of projects received in the past is detected and called P1

4. Calculate "Average value of P1 = sum of project expenses in P1 / number of projects in P1"

5. As the absolute value of the difference between the project cost of the recommended project and the average value of P1 is closer to 0, the highest score is given.

<Business type (share, ratio)>

The proportion (ratio) by business type is determined by using the proportion of each business type that has been ordered and the ranking of the business type searched by the user (company) from past history data of the user (company).

The business type determining unit 402 analyzes the proportion of each business type of the existing projects, and the proportion of each business type based on the total order received (article: construction: service), the proportion of each business type based on the business received in the last 6 months (article: construction: service) derive

In addition, the business type determining unit 402 derives the proportion (product: construction: service) of each business type searched by the user (company) for the last one month.

The evaluation value calculation unit 406 for each project element derives a recommended weight for each project type by adjusting the deviation of the derived weight for each project type (product: construction: service). For example, it can be 10% of goods, 10% of construction, and 80% of services (10:10:80).

<Detailed business type>

The detailed business type is determined by using the keywords of the existing business order and the keywords searched by the user (company) from the user's (company)'s past history data.

The detailed business type determination unit 403 analyzes the keywords of the existing order-taking projects, processes stop words such as place names, organization names, and projects, and detailed business types based on keywords through ranking and relationship analysis of the first keywords to derive

In addition, the detailed business type determining unit 403 derives the ranking of the second keyword searched by the user (company) for the last one month.

The evaluation value calculation unit 406 for each business element adjusts the recommendation order of keyword criteria in the order of the degree of overlap (congruence) within the upper keyword range in the ranking of the first keyword and the ranking of the second keyword based on the derived keywords. do.

For example, based on the search date (current) or the last access date, the evaluation value for the detailed business type is calculated according to the order of mapping with the upper keyword among new announcements whose notification registration date is today or after the last access date.

Here, the evaluation value for the detailed business type is calculated in the same order as the keyword overlapping order of the corresponding company in the case of a business including a keyword with a degree of overlap. At this time, the evaluation value is calculated so that the 1st priority is the highest point and the ranking becomes lower as it goes down. In addition, if a plurality of keywords are included in one business, additional points are given.

That is, the evaluation value (keyword hit evaluation value) for the detailed business type can be calculated as follows.

1. Keyword analysis of the company

2. Detection and scoring of duplicate keyword sequences of the first keyword

3. Detect duplicate keyword sequence of second keyword and give score

4. Weighting of duplicate keywords of the second keyword (the ratio of the first keyword in the total number of keywords removed from duplicates)

5. The company's final keyword ranking and score

6. Keyword extraction and analysis for each recommended business

7. Points are given to the corresponding keywords (additional points are given for multiple keywords)

<Customer Type>

For reference, each institution according to the customer type (institution type) is listed in FIG. 6 .

The customer type is determined using the existing customers (organizations), the proportion of existing customers (organizations) by type, and the customer ranking searched by the corresponding user (company).

The customer type determining unit 404 analyzes the institutional types of existing customers by deriving the proportion of each institution type based on the total number of orders received or by deriving the proportion of each institution type based on the projects received in the last year.

In addition, the customer type determining unit 404 determines the proportion (ratio) of the type of ordering organization and the type of ordering organization searched for by the corresponding company for the last 6 months (see details).

The evaluation value calculation unit 406 for each project element derives the weight of the ordering organization type when recommending the project by adjusting the deviation of the determined weight for each type of organization. For example, the proportion of the type of ordering institution may be 50% of autonomous administrative institutions, 30% of national administrative institutions, 12% of educational administrative institutions, and 8% of public institutions (50:30:12:8).

<Business Location>

The business location is determined using the rank of the location of the company, the location of the existing customer, and the location of the customer searched by the company.

The business location determining unit 405 derives the distance ranking between the corresponding company and existing customers based on the unit of the local government.

In addition, the business location determining unit 405 derives a distance ranking between the corresponding company and the ordering agencies (customers) searched for by the corresponding company based on the unit of the local government.

The evaluation value calculation unit 406 for each business element calculates a distance index with the corresponding company based on the derived list of ordering organizations, and recommends a project according to the distribution ratio for each distance index.

Here, the distance index is the absolute value of the difference between the location code of the company and the location code of the business, and the distance index can be calculated as how many boundaries must pass based on the administrative district.

For example, if a company and business are located in the same Gyeonggi-do, the distance index may be 0, and if the company and business are located in Gyeonggi-do and Chungcheong-do, respectively, the distance index may be 1. Accordingly, the distribution weight for each distance index is 50% for distance index "0", 30% for distance index "1", 10% for distance index "2", 5% for distance index "3", 5% for distance index "5". can decide

That is, when recommending a business, the evaluation value for the business location can be evaluated as follows.

In calculating the evaluation value for the distance between the company and the ordering agency (business location) by applying the modeled distance index between the local governments, it can be calculated as follows.

1. If the distance index is 0, the highest point is given

2. Using the distance index calculation formula between the company and the business

Distance index = ㅣ Company location code ?? Business location codeㅣ (absolute value of location code difference)

5 is a detailed configuration diagram of the artificial intelligence analysis unit of FIG. 2 .

As shown in FIG. 5 , the artificial intelligence analysis unit 105 according to the present invention includes a business announcement recommendation unit 501 , a bidding result analysis unit 502 , a hit rate calculation unit 503 for each element, and a learning unit 504 . ), an inference unit 505 , a weight modeling unit 506 , and a weight adjustment unit 507 for each element.

The business announcement recommendation unit 501 recommends a business announcement according to the sum total using the evaluation value for each element of the business and the weight for each element for the target business.

For example, the initial weight for each of the five elements may be equally given by 20% (0.2). Thereafter, the weight for each element may be corrected through the weight correction step.

The bidding result analysis unit 502 analyzes the bidding result based on the user's satisfaction feedback for the recommended project announcement and the corresponding bidding result database 106e.

The hit ratio calculation unit 503 for each element calculates a hit ratio for each analysis element from the bidding result.

The learning unit 504 learns the characteristics of the user's (company) winning business based on the calculated hit ratio and the hit ratio trend according to the past history.

The reasoning unit 505 extracts and infers factors related to successful bids from the learned characteristics of the order receiving business.

The weight modeling unit 506 models and updates a weighting method for each analysis element according to the inference result.

The weight adjustment unit 507 for each element adjusts a weight for each analysis element according to the modeled weighting method.

The artificial intelligence analysis unit 105 further includes an initial weight determination unit 508 for determining an initial weight of the user based on the weight model database 106f and individual characteristics of the user.

6 is a view for explaining the customer types classified in the bidding strategy establishment support system using the personalization-based business recommendation algorithm according to the present invention.

Referring to FIG. 6 , customer types can be classified into public periods, educational institutions, educational administrative institutions, national administrative institutions, committees, legislative institutions, autonomous administrative institutions, constitutional institutions, and the like.

7 is a flowchart illustrating a method for supporting establishment of a bidding strategy using a personalization-based business recommendation algorithm according to an embodiment of the present invention.

First, data necessary for establishing a bidding strategy are collected from a plurality of external data servers 130 and 140 (S710).

Thereafter, the user's past business history information is acquired and processed from the collected data to determine the business recommendation range and criteria including the priority and weight of each element to which the user's individual characteristics are applied (S720).

Thereafter, an evaluation value for each element of the target project included in the project recommendation range and standard is calculated (S730).

Thereafter, the project announcement is recommended according to the total score summed using the evaluation value for each element of the project for the target project and the weight for each element (S740).

For example, the initial weight for each of the five elements may be equally given by 20% (0.2). Thereafter, the weight for each element may be corrected through the weight correction step.

Thereafter, the characteristics of each element according to the user's satisfaction feedback and the bidding result are analyzed with respect to the recommended business announcement (S750).

Thereafter, the weight for each element is corrected by learning the hit ratio for each element and the transition learning of the hit ratio for each element (S760).

Thereafter, the project recommendation step (S740), the element-specific characteristic analysis step (S750), and the weight correction step (S760) are repeatedly performed to provide a bidding strategy (S770).

8 is a detailed flowchart of an embodiment of the data collection step ( S710 ) of FIG. 7 .

The data collection step (S710) is. First, structured data and unstructured data are collected from a plurality of external data servers 130 and 140 (S810).

Thereafter, the collected data is processed (S820).

Thereafter, keywords are extracted from the processed data (S830).

Thereafter, the extracted keyword is input to the database (S840).

Thereafter, the input signal of the user terminals 110 and 120 is received and converted (S850).

Thereafter, a characteristic element is extracted from the converted input signal of the user terminals 110 and 120 ( S860 ).

9 is a detailed flowchart of an embodiment of the data processing step ( S720 ) of FIG. 7 .

In the data processing step (S720), first, the business scale of the business announcement to be recommended to the user (company) is determined (S910).

A business type of the business announcement to be recommended to the user is determined (S920).

A detailed business type of the business announcement to be recommended to the user is determined (S930).

A customer type of the business announcement to be recommended to the user is determined (S940).

The business location of the business announcement to be recommended to the user is determined (S950).

Thereafter, the business recommendation scope and criteria are determined based on the determined business scale, business type, detailed business type, customer type, and business location (S960).

The project size determination step (S910) determines the project size as a preset amount range by calculating the distribution, average value, and trend of each type of the existing project, and the project type determination step (S920) is the existing implementation The proportion (ratio) of the business type is determined by preference according to the business type ratio of "Service: Goods: Construction" of the project, and the detailed business type determination step (S930) is through the analysis of keywords in the public notice of the existing project. , the detailed business type is determined, and the customer type determination step (S940) determines the proportion (ratio) of the customer type with preference according to the customer type of the existing business, and the business location determination step (S950) includes, It is determined using the location (location) of the user (company) and the business location of the customer (ordering organization).

In the element-by-element evaluation value calculation step (S730), a determination method for each element will be described in detail.

<Business scale (public notice project cost)>

The business scale is determined by using the project cost for each type of the existing order received from the user (company)'s past history data and the project cost of the business that the user (company) has inquired in detail (link moves).

In the project size determination step (S910), the project cost for each project type of the existing order-based project is analyzed with respect to the maximum project cost by type, the minimum project cost by type, the distribution and section of the project cost by type, and the trend of the average project cost by type by year/month.

In addition, in the business size determination step (S910), the distribution map and section of the project cost by type of business that the user (company) has searched-linked in the last 3 months and the distribution of the project cost by type of the business that the user (company) has searched-linked in the last 1 month Analyzes the business cost of the business that the user (company) inquires (moves the link) in detail for the section and the like.

In the element-specific evaluation value calculation step (S730), the project recommendation range is derived by adjusting the upper limit for each type, the lower limit for each type, the upper/lower limit of the distribution weight 1st priority section, and the recommended section according to the average movement trend.

In addition, in the element-specific evaluation value calculation step (S730), the average value of the 1st priority section of the search-link moving business criteria distribution for the last 3 months, the 1st rank section of the distribution ratio of the business criteria that the user (company) has searched-linked for the last 1 month The scope of the project recommendation is reduced by deriving the average value of

Here, if the average movement trend is on the rise, the closer to the average value in the project range equal to or higher than the average value of the 1st priority section, the higher the ranking. The closer to , the higher the ranking, and the evaluation value for the business size is given. At this time, the evaluation value is calculated so that the 1st priority is the highest point and the ranking becomes lower as it goes down.

Here, the evaluation value for the business size can be calculated as follows.

1. Classification of sections by a preset amount based on the company’s past history

2. Count the number of projects for each section

3. The project cost section in which 70% of the total number of projects received in the past is detected and called P1

4. Calculate "Average value of P1 = sum of project expenses in P1 / number of projects in P1"

5. As the absolute value of the difference between the project cost of the recommended project and the average value of P1 is closer to 0, the highest score is given.

<Business type (share, ratio)>

The proportion (ratio) by business type is determined by using the proportion of each business type that has been ordered and the ranking of the business type searched by the user (company) from past history data of the user (company).

In the project type determination step (S920), the proportion of each business type of the existing projects is analyzed, and the proportion of each business type based on the total orders received (product: construction: service), and the proportion of each business type based on the business received in the last 6 months (product: construction: service) derive

In addition, in the business type determination step ( S920 ), the proportion (product: construction: service) of each business type searched by the user (company) for the last one month is derived.

In the element-specific evaluation value calculation step (S730), the recommended weight for each project type is derived by adjusting the deviation of the derived weight for each type of project (product: construction: service). For example, it can be 10% of goods, 10% of construction, and 80% of services (10:10:80).

<Detailed business type>

The detailed business type is determined by using the keywords of the existing business order and the keywords searched by the user (company) from the user's (company)'s past history data.

In the detailed business type determination step (S930), keywords of the existing project to be won are analyzed, stop words such as place names, organization names, and projects are processed, and detailed business types based on keywords through ranking and relationship analysis of the first keyword to derive

In addition, in the detailed business type determination step (S930), the ranking of the second keyword searched by the user (company) for the last one month is derived.

In the element-specific evaluation value calculation step (S730), the keyword recommendation order is adjusted in the order of the degree of overlap (congruence) within the upper keyword range in the ranking of the first keyword and the ranking of the second keyword based on the derived keywords. .

For example, based on the search date (current) or the last access date, the evaluation value for the detailed business type is calculated according to the order of mapping with the upper keyword among new announcements whose notification registration date is today or after the last access date.

Here, the evaluation value for the detailed business type is calculated in the same order as the keyword overlapping order of the corresponding company in the case of a business including a keyword with a degree of overlap. At this time, the evaluation value is calculated so that the 1st priority is the highest point and the ranking becomes lower as it goes down. In addition, if a plurality of keywords are included in one business, additional points are given.

That is, the evaluation value (keyword hit evaluation value) for the detailed business type can be calculated as follows.

1. Keyword analysis of the company

2. Detection and scoring of duplicate keyword sequences of the first keyword

3. Detect duplicate keyword sequence of second keyword and give score

4. Weighting of duplicate keywords of the second keyword (the ratio of the first keyword in the total number of keywords removed from duplicates)

5. The company's final keyword ranking and score

6. Keyword extraction and analysis for each recommended business

7. Points are given to the corresponding keywords (additional points are given for multiple keywords)

<Customer Type>

For reference, each institution according to the customer type (institution type) is listed in FIG. 6 .

The customer type is determined using the existing customers (organizations), the proportion of existing customers (organizations) by type, and the customer ranking searched by the corresponding user (company).

In the customer type determination step (S940), the institutional type of existing customers is analyzed by deriving the proportion of each institution type based on the total number of orders received or by deriving the proportion of each institution type based on the projects received in the last year.

In addition, in the customer type determination step (S940), the proportion (ratio) of the type of ordering organization and the type of ordering organization searched for by the corresponding company for the last 6 months is determined.

In the element-specific evaluation value calculation step (S730), the weight of the ordering organization type at the time of project recommendation is derived by adjusting the deviation of the determined weight for each type of organization. For example, the proportion of the type of ordering institution may be 50% of autonomous administrative institutions, 30% of national administrative institutions, 12% of educational administrative institutions, and 8% of public institutions (50:30:12:8).

<Business Location>

The business location is determined using the ranking of the location of the company, the location of the existing customer, and the location of the customer searched for by the company.

In the business location determination step (S950), a distance ranking between the corresponding company and existing customers is derived based on the unit of the local government.

In addition, in the project location determination step (S950), the distance ranking between the corresponding company and the ordering agencies (customers) searched for by the corresponding company is derived based on the unit of the local government.

In the element-specific evaluation value calculation step (S730), a distance index with the corresponding company is calculated based on the derived list of ordering organizations, and a project is recommended according to the distribution ratio of each distance index.

Here, the distance index is the absolute value of the difference between the location code of the company and the location code of the business.

For example, if a company and business are located in the same Gyeonggi-do, the distance index may be 0, and if the company and business are located in Gyeonggi-do and Chungcheong-do, respectively, the distance index may be 1. Accordingly, the distribution weight for each distance index is 50% for distance index "0", 30% for distance index "1", 10% for distance index "2", 5% for distance index "3", 5% for distance index "5". can decide

That is, when recommending a business, the evaluation value for the business location can be evaluated as follows.

In calculating the evaluation value of the distance between the company and the ordering agency (business location) by applying the modeled distance index between the local governments, it can be calculated as follows.

1. If the distance index is 0, the highest point is given

2. Using the distance index calculation formula between the company and the business

Distance index = ㅣ Company location code ?? Business location codeㅣ (absolute value of location code difference)

10 is a detailed flowchart of an embodiment of the weight correction step S760 of FIG. 7 .

In the weight correction step ( S760 ), first, a hit ratio for each element for each analysis element is calculated from the bidding result ( S1010 ).

Here, the hit ratio for each element is calculated as follows.

The hit rate of factor A can be calculated as the number of projects that include factor A among the total number of thousands of projects.

Then, based on the calculated hit ratio and the hit ratio trend according to the past history, the characteristics of the user's (company) order receiving business are learned ( S1020 ).

Thereafter, factors related to successful bids are extracted and inferred from the learned characteristics of the order receiving business (S1030).

Thereafter, a weighting method for each analysis element is modeled or updated according to the inference result ( S1040 ).

Thereafter, the weight for each analysis element is adjusted (S1050).

Although the method for supporting establishment of a bidding strategy using a personalization-based business recommendation algorithm according to an embodiment of the present invention has been described above, a computer-readable record storing a program for implementing a method for supporting the establishment of a bidding strategy using a personalization-based business recommendation algorithm It goes without saying that a program stored in a computer-readable recording medium for implementing a method for supporting establishment of a bidding strategy using a medium and a personalization-based business recommendation algorithm can also be implemented.

That is, those skilled in the art can easily understand that the above-described method for supporting establishment of a bidding strategy using the above-described personalization-based business recommendation algorithm may be provided by being included in a computer-readable recording medium by tangibly implementing a program of instructions for implementing it. You will understand. In other words, it may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the computer-readable recording medium may be specially designed and configured for the present invention, or may be known and used by those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and floppy disks. magneto-optical media, and hardware devices specially configured to store and carry out program instructions, such as ROM, RAM, flash memory, USB memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the gist of the present invention as claimed in the claims.

100: bid strategy establishment support server
110: first user terminal 120: second user terminal
130: external data server 1 140: external data server N
101: interface unit 102: control unit
103: data collection unit 104: data processing unit
105: artificial intelligence analysis unit 106: database management unit
301: collection unit 302: processing unit
303: keyword extraction unit 304: DB input unit
305: data conversion unit 306: characteristic element extraction unit
401: business size determining unit 402: business type determining unit
403: detailed business type determining unit 404: customer type determining unit
405: business location determination unit 406: evaluation value calculation unit for each business element
501: business announcement recommendation unit 502: bidding result analysis unit
503: hit rate calculation unit 504: learning unit
505: inference unit 506: weight modeling unit
507: element weight adjustment unit 508: initial weight determination unit
S710: data collection step
S720: data processing step
S730: Evaluated value calculation step for each element
S740: Business notification recommendation stage
S750: Element-specific characteristic analysis step step
S760: weight correction step
S770: Bidding strategy provision step

Claims (10)

In the bidding strategy establishment support system using the personalization-based business recommendation algorithm,
By using the business recommendation algorithm according to the individual characteristics of the user, the weight of each analysis element is corrected through the learning of the hit rate and the hit rate trend by the analysis element by interaction with the user to recommend the relevant bid announcement and to establish a customized bidding strategy Bidding strategy collection support server 100 for;
a plurality of user terminals (110, 120) for accessing the bidding strategy collection support server, transmitting and receiving data necessary to receive a customized business announcement, and receiving a customized business announcement according to the business recommendation algorithm; and
A plurality of external data servers 130 and 140 for collecting data necessary to establish and provide a bidding strategy according to the customized business announcement
including,
The bidding strategy collection support server 100,
an interface unit 101 for providing an interface for providing the customized business announcement to the plurality of user terminals;
a data collection unit 103 for collecting and processing data from the plurality of external data servers, extracting keywords, inputting them into a database, and extracting individual characteristic elements from an input signal through the user terminal;
a data processing unit 104 for performing data processing necessary to provide the customized business announcement based on data transmitted and received through the interface unit and data collected through the data collection unit;
an artificial intelligence analysis unit 105 for updating the weight model by analyzing the hit ratio and the hit ratio trend according to the selection input of the user terminal and the bidding result;
a database management unit 106 that stores data necessary to provide the customized business announcement; and
Control unit 102 for controlling each component including the interface unit, data collection unit, data processing unit, artificial intelligence analysis unit, and database management unit
A bidding strategy establishment support system using a personalization-based business recommendation algorithm, characterized in that it comprises a.
According to claim 1,
The data collection unit 103,
a collecting unit 301 for collecting structured data and unstructured data from the plurality of external data servers;
a processing unit 302 for processing the collected data;
a keyword extraction unit 303 for extracting keywords from the processed data;
a DB input unit 304 for inputting the extracted keyword into a database;
a data conversion unit 305 for receiving and converting the input signal of the user terminal; and
A user characteristic extracting unit 306 for extracting a user characteristic element from the converted input signal of the user terminal
A bidding strategy establishment support system using a personalization-based business recommendation algorithm, characterized in that it comprises a.
According to claim 1,
The data processing unit 104,
a business scale determining unit 401 for determining a business scale of a business announcement to be recommended to the user from the data stored in the database;
a business type determination unit 402 for determining a business type of a business announcement to be recommended to the user from the data stored in the database;
a detailed business type determination unit (403) for determining a detailed business type of a business announcement to be recommended to the user from the data stored in the database;
a customer type determining unit 404 for determining a customer type of a business advertisement to be recommended to the user from the data stored in the database;
a business location determination unit (405) for determining a business location of a business announcement to be recommended to the user from the data stored in the database; ??
A project for determining the business recommendation scope and criteria based on the determined business scale, business type, detailed business type, customer type and business location, and calculating an evaluation value for each element of the target business included in the business recommendation scope and criteria Evaluation value calculation unit 406 for each element
A bidding strategy establishment support system using a personalization-based business recommendation algorithm, characterized in that it comprises a.
4. The method of claim 3,
The artificial intelligence analysis unit 105,
a business announcement recommendation unit 501 for recommending a business announcement according to a total score summed using the evaluation value for each element of the business for the target business and the weight for each element;
a bidding result analysis unit 502 for analyzing the bidding result based on the user's satisfaction feedback for the recommended project announcement and the following bidding result database;
a hit ratio calculation unit 503 for calculating a hit ratio for each analysis element from the bidding result;
a learning unit 504 for learning the characteristics of the user's (company) order receiving business based on the calculated hit ratio and the hit ratio trend according to the past history;
an inference unit 505 for extracting and inferring factors related to successful bids from the learned characteristics of the order receiving business;
a weight modeling unit 506 for modeling a weighting method for each analysis element according to the inference result; and
Weight adjustment unit 507 for each element for adjusting the weight for each analysis element
A bidding strategy establishment support system using a personalization-based business recommendation algorithm, characterized in that it comprises a.
5. The method of claim 4,
The business scale determining unit 401,
It is characterized in that the size of the project is determined in a preset amount range by calculating the distribution, average value, and trend of each type of existing project,
The business type determination unit 402,
It is characterized in that the proportion (ratio) of the business type is determined by preference according to the business type ratio of "service: article: construction" of the existing project,
The detailed business type determination unit 403,
It is characterized in that the detailed project type is determined through the keyword analysis of the public notice of the existing project,
The customer type determining unit 404,
It is characterized in that the proportion (ratio) of the customer type is determined by preference according to the customer type of the existing business,
The business location determining unit 405,
It is characterized in that it is determined using the location (location) of the user (company) and the business location of the customer (ordering organization),
The evaluation value calculation unit 406 for each business element,
With respect to the business scale, the user (company) divides the section by a preset amount based on the history of the user (company), counts the number of projects for each section, and detects the project cost section in which 70% of the total number of projects received in the past is distributed. Let's say P1, calculate "average value of P1 = sum of project expenses in P1 / number of projects in P1" characterized in that it is calculated,
With respect to the above business type, it is characterized in that the recommended weight for each project type is derived by adjusting the deviation of the derived weight for each project type (product: construction: service),
For the detailed business type, the keywords of the corresponding company are analyzed, the duplicate keyword sequence detection and score of the first keyword of the existing business order are detected and the duplicate keyword sequence of the second keyword searched by the user (company) for the last 1 month Detection and scoring, weighting the duplicated keywords of the second keyword, giving the company's final keyword ranking and score, extracting and analyzing keywords for each recommended business, and assigning scores to the keywords; , characterized in that the evaluation value is calculated by adding additional points to a plurality of keywords,
With respect to the customer type, it is characterized in that the proportion of the ordering organization type at the time of business recommendation is derived by adjusting the deviation of the determined proportion (ratio) of the customer type,
With respect to the business location, a bidding strategy establishment support system using a personalization-based business recommendation algorithm, characterized in that calculating a distance index that is an absolute value of the difference between the location code of the company and the location code of the business.
A method for supporting establishment of a bidding strategy using a personalization-based business recommendation algorithm in the form of a program executed by a calculation processing means including a computer, the method comprising:
A data collection step (S710) of collecting data necessary for establishing a bidding strategy from a plurality of external data servers;
A data processing step (S720) of acquiring and processing the user's past business history information from the collected data to determine the business recommendation range and criteria including the priority and weight of each element to which the user's individual characteristics are applied (S720)
a factor-by-element evaluation value calculation step (S730) of calculating an evaluation value for each element of the target project included in the project recommendation range and standard;
a business announcement recommendation step (S740) of recommending a business announcement according to a total score summed using the evaluation value for each element of the business and the weight for each element for the target business;
element-specific characteristic analysis step (S750) of analyzing element-specific characteristics according to the user's satisfaction feedback and bidding results for the recommended project announcement;
a weight correction step (S760) of correcting the weight for each element by learning the hit ratio for each element and the transition learning of the hit ratio for each element; and
A bidding strategy providing step of providing a bidding strategy by repeating the project recommendation step, the element-specific characteristic analysis step, and the weight correction step (S770)
A method of supporting the establishment of a bidding strategy using a personalization-based business recommendation algorithm comprising a.
7. The method of claim 6,
The data collection step (S710) is,
a collecting step of collecting structured data and unstructured data from the plurality of external data servers (S810);
A processing step of processing the collected data (S820);
A keyword extraction step of extracting a keyword from the processed data (S830);
DB input step of inputting the extracted keyword into the database (S840);
a data conversion step of receiving and converting the input signal of the user terminal (S850); and
A characteristic element extraction step of extracting characteristic elements from the converted input signal of the user terminal (S860)
A method of supporting the establishment of a bidding strategy using a personalization-based business recommendation algorithm, characterized in that it comprises a.
7. The method of claim 6,
The data processing step (S720) is,
a business scale determination step (S910) of determining the business scale of the business announcement to be recommended to the user;
a business type determination step (S920) of determining the business type of the business announcement to be recommended to the user;
a detailed business type determination step of determining the detailed business type of the business announcement to be recommended to the user (S930);
a customer type determination step (S940) of determining a customer type of a business advertisement to be recommended to the user;
a business location determination step (S950) of determining a business location of a business announcement to be recommended to the user; and
Recommendation range determination step (S960) of determining the business recommendation scope and criteria based on the determined business scale, business type, detailed business type, customer type and business location
A method of supporting the establishment of a bidding strategy using a personalization-based business recommendation algorithm, characterized in that it comprises a.
7. The method of claim 6,
The business scale determination step (S910) is,
It is characterized in that the size of the project is determined in a preset amount range by calculating the distribution, average value, and trend of each type of existing project,
The business type determination step (S920) is,
It is characterized in that the proportion (ratio) of the business type is determined by preference according to the business type ratio of "service: article: construction" of the existing project,
The detailed business type determination step (S930) is,
It is characterized in that the detailed project type is determined through the keyword analysis of the public notice of the existing project,
The customer type determination step (S940) is,
It is characterized in that the proportion (ratio) of the customer type is determined by preference according to the customer type of the existing business,
The business location determination step (S950) is,
It is characterized in that it is determined using the location (location) of the user (company) and the business location of the customer (ordering organization),
The element-specific evaluation value calculation step (S730) is,
With respect to the business scale, the user (company) divides the section by a preset amount based on the history of the user (company), counts the number of projects for each section, and detects the project cost section in which 70% of the total number of projects received in the past is distributed. Let's say P1, calculate "average value of P1 = sum of project expenses in P1 / number of projects in P1" characterized in that it is calculated,
With respect to the above business type, it is characterized in that the recommended weight for each project type is derived by adjusting the deviation of the derived weight for each project type (product: construction: service),
For the detailed business type, the keywords of the corresponding company are analyzed, the duplicate keyword sequence detection and score of the first keyword of the existing business order are detected and the duplicate keyword sequence of the second keyword searched by the user (company) for the last 1 month Detection and scoring, weighting the duplicated keywords of the second keyword, giving the company's final keyword ranking and score, extracting and analyzing keywords for each recommended business, and assigning scores to the keywords; , characterized in that the evaluation value is calculated by adding additional points to the plurality of keywords,
With respect to the customer type, it is characterized in that the proportion of the ordering organization type at the time of business recommendation is derived by adjusting the deviation of the determined proportion (ratio) of the customer type,
For the business location, calculate the distance index, which is the absolute value of the difference between the location code of the company and the location code of the business
A method of supporting the establishment of a bidding strategy using a personalization-based business recommendation algorithm, characterized in that.
7. The method of claim 6,
The weight correction step (S760) is,
a hit ratio calculation step of calculating a hit ratio for each element for each analysis element from the bidding result (S1010);
a learning step (S1020) of learning the characteristics of the user (company)'s order receiving business based on the calculated hit ratio and the hit ratio trend according to the past history;
an inference step (S1030) of extracting and inferring factors related to successful bids from the learned characteristics of the order receiving business;
a weight modeling step of modeling a weighting method for each analysis element according to the inference result (S1040); and
Adjustment step of adjusting the weight for each analysis element (S1050)
A method of supporting the establishment of a bidding strategy using a personalization-based business recommendation algorithm, characterized in that it comprises a.

Images