ROLE-BASED PROPOSAL DECISION JUSTIFICATION SYSTEM

Abstract

This invention relates to a system (1) that justifies bid evaluation decisions in electronic tendering processes according to user role and improves the production of explanations in line with user feedback.

Description

1 TARIFF ROLE-BASED PROPOSAL DECISION JUSTIFICATION SYSTEM Technical Area This invention enables bids to be submitted using artificial intelligence in e-auction applications. each It provides natural language descriptions specific to the user's role and receives feedback from the user. Improving the explanation structure with notifications while creating a next-generation decision support mechanism. It is related to a system that provides. 10 Previous Technique In the known state of the art, bids are determined on electronic tendering platforms. 15 using rules or AI-powered evaluation algorithms Scoring and suitability assessments are carried out. Current In systems, acceptance or rejection decisions resulting from evaluation are usually made. scores, rankings, or pre-prepared static text are provided to the user. It is reported. In known solutions, which technical or operational decision is made? It was created based on reasons and provided to the user with sufficient detail. 20 It cannot be explained. The explanations provided are mostly the same for all users. The content is being prepared based on the user's technical knowledge level or system. It is not differentiated according to its role within the organization. As a result... The comprehensibility of decisions is decreasing, and objections to decision-making processes are increasing. and user trust is negatively affected. Therefore, for users, a further 25 a new generation decision support mechanism aiming to provide detailed explanations It is needed. United States Regulation US20170286870A1, which is included in the known state of the art. In his patent, he describes the collection, analysis of bids and the selection of the winning 30 in tender processes. 2 a system that manages processes such as proposal selection using artificial intelligence It is mentioned. Brief Description of the Invention The purpose of this invention is to improve the bid evaluation process used in electronic tendering procedures. a decision statement that explains the reasons for its decisions to the user in an understandable way The goal is to provide the system. Another aim of the invention is to provide the technical details needed by different user roles. It is about dynamically changing the description content according to its level. Another aim of the invention is to increase user satisfaction and by analyzing comprehensibility feedback, the explanation production mechanism The goal is to ensure its continuous improvement. 15 Detailed Description of the Invention The "Role-Based Proposal Decision" was implemented to achieve the purpose of this invention. The "Justification System" is shown in the attached document; this document is numbered 20. Figure 1. Schematic view of a system that is the subject of the invention. The parts shown in the figure are individually numbered, and these numbers correspond to... The corresponding answers are given below. 25 1. System 2. Electronic device 3. Database 4. Server 30 D. External server 3 The bid evaluation results generated in electronic tendering processes explaining to the user and providing explanations based on user feedback. The system that is the subject of the invention, which is constantly improving its quality (1); - User enters the request for quotation and quotation data. 5 providing an interface that allows remote access via a web browser or application. at least one electronic device capable of communicating with servers (2), - price, delivery time, technical qualification information, to suppliers or Customer offer details, product / service descriptions, prices at least one 10 configured to store information and historical offers database (3) and - machine learning based on data taken from the database (3) or Scoring bids using rule-based algorithms helps select the most suitable bids. prioritizing, scoring bids according to existing rules to perform a suitability assessment by matching them, and to reject unsuitable bids. 15 filtering and transferring to the explanation module, the role of the logged-in user to identify and tailor the description and interface the user will see to this role. personalized offers, low-scoring offers, or offers that do not comply with the rules to provide the user with an explanation of cause and effect, what is done 20 explanations depending on the technical level the user can understand to simplify or elaborate on, the same explanations offering it in different languages, supporting international users. to provide the generated outputs to the user in a visual and interactive way to present, from the user’s interactions on the electronic device (2) interface By extracting meaning and analyzing system performance, feedback is provided. 25 to be created and to become more effective over time through the feedback received, to enable it to provide personalized and understandable explanations It includes at least one configured server (4). The system in question consists of (1) electronic devices (2) smartphones, tablet computers, 30 It is a device similar to a desktop computer or a portable computer. The one in question... 4 electronic device (2) any remote device in the known state of the art to connect to the server (4) using the communication protocol is structured. The preferred application of the invention is an electronic device (2) exchanging data with the server (4) using the internet as a data bus It is structured as follows: Electronic device (2), offer information, specification 5 criteria, evaluation parameters and user-related processes It is configured to provide a user interface that allows input. Offer evaluation generated by electronic device (2), server (4) results, reasoning behind decisions, natural language customized according to user role explanations and multilingual explanation outputs are presented to the user in graphical and textual formats. It is structured to ensure that it is presented as such. In the system that is the subject of the invention, the database (3) in (1) is in communication with the server (4) and is configured to be managed by the server (4). The invention is preferred In the arrangement made, the database (3) contains the offer price information, delivery time 15 information, technical suitability data, specification clauses, offer evaluation scores, user role information, user authorization levels, evaluation results, reasoning behind decisions, natural language explanatory texts, role-based description templates, multilingual description content, user satisfaction data, comprehensibility assessments, user feedback 20 notifications, explanation learning parameters, artificial intelligence model parameters, description, production history, system transaction logs, user transaction records, log records, and configuration parameters It is configured to ensure that it is saved. The server (4) in the system (1) which is the subject of the invention, any remote communication to communicate with the electronic device (2) using the protocol and the established protocol to exchange data with electronic devices (2) via communication is being configured. The server (4) in question enters new data into the database (3). registration, deletion of registered data in database (3) or data 30 Modifying the registered data in the database (3), database (3), within Data processing involves operations such as updating and retrieving recorded data. It is structured to manage the base (3). The preferred method of the invention In the application, the server (4) is activated by the user via the electronic device (2). to receive the request for evaluation of bids in the submitted e-bidding processes is being configured. Server (4), bid evaluation in e-tender applications 5 It is configured to start the process. The server (4) receives the offer data, price information, delivery time information, technical compliance data, specifications criteria in the form of a database (3), cloud server or enterprise information external It is configured to receive the offer via server (D). Server (4) receives the offer. using data evaluation rules and machine learning algorithms 10 It is structured to enable analysis. The server (4), offer calculating scores, evaluating the conformity of bids to the specifications. to carry out and evaluate the decision reasons that result from the evaluation. It is structured to determine the server (4), in the bidding processes. Identifying the user's role and creating 15 based on the defined user role. the level of detail, technical content density, and style of expression of the statement It is configured to determine the server (4), the user's supplier, manager, Which of the different user roles does it belong to, such as purchasing specialist or auditor? to determine this and the level of detail and technical content of the statement to be created. It is structured to determine its level and mode of expression according to this role. 20 The server (4) presents the bid evaluation results, scores, specification discrepancies and to create natural language explanations using the reasoning behind the decision It is structured. Server (4), natural language generation (Natural Language Generation- It is structured to generate explanatory texts using the NLG (Network Language Generation) method. Server (4) generates descriptions, role-based language customization mechanism 25 to simplify, elaborate on, or change the technical level through It is structured in such a way. The server (4) uses ontology instead of custom language libraries. using terminology matching based on dictionary definitions or jargon filtering based on dictionary definitions It is structured as follows: Server (4), role-based language customization, ontology terminology matching based on vocabulary, dictionary-based phrase filtering, language rules 30 This can be achieved using text processing methods such as: 6 It is structured. The server (4) describes the large language model (Large Language Model-LLM), Sequence-to-Sequence-Seq2Seq learning model, Transformer-based models or algorithms It is configured to produce with hybrid structures in which it is used together. Server (4), generated descriptions are machine translated (MT) 5 using methods or multilingual description templates to translate into different languages to convert and enable international users to receive descriptions in their own languages It is configured to provide the server (4), multilingual conversion process, machine translation models, multilingual text templates, or external language conversion It is configured to provide explanations using the services. Server (4), 10 User satisfaction transmitted via electronic device (2) their evaluations, comprehensibility scores, competency assessments, and The server (4) is configured to receive textual feedback. notifications using machine learning (ML) algorithms to analyze, explain, update the parameters of the production engine and 15 The explanation is structured to improve production quality. Server (4), in carrying out the learning process for improvement (Supervised Learning), active learning or reinforcement learning Q-learning (Q-RL) in the reinforcement learning approach with methods The learning algorithm is configured for use. Server (4), 20 the results of the proposal evaluation, the reasons for the decision, and the explanation texts, user feedback, description generation model parameters, by storing learning outcomes and transaction records on the database (3) It is configured to ensure traceability. The server (4) is in the database. (3) continuous learning and explanation of information improves production performance over time 25 It is structured to ensure its improvement. 7 Industrial Application of the Invention Historical and current bid data, evaluation scores, technical suitability information, specification criteria, user roles and user feedback in natural language processing, AI-based explanation generation, machine learning algorithms, and The subject of the invention is 5, thanks to its analysis using dynamic decision-making mechanisms. system (1) directly integrated into electronic tendering, purchasing, supply management This can be done. Thus, manual decision-making processes are reduced, and proposals are made available. The transparency and accountability of evaluation processes are being increased, and appeals and Evaluation costs are reduced, user satisfaction is increased, and Electronic bidding increases the efficiency of procurement operations. 10 Around these fundamental concepts, the invention focuses on "Role-Based Proposal Decision Making". Development of various regulations regarding the Justification System (1)" It is possible, and the invention is not limited to the examples described here, but is essentially As stated in the requests. 15

Claims

8 REQUESTS 1. The bid evaluation results generated in electronic tendering processes. explaining to the user and providing explanations based on user feedback. continuously improving its quality; - Used by the user, for entering offer data. 5 a web browser that provides an interface for performing operations or at least one that can communicate with remote servers via the application electronic device (2), - price, delivery time, technical qualification information, to suppliers or Customer offer details, product / service descriptions, price 10 at least one configured to store information and historical offers database (3) and - machine learning based on data taken from the database (3) or Scoring bids using rule-based algorithms helps select the most suitable bids. To prioritize, scored bids will be evaluated according to existing rules (15). to perform a suitability assessment by matching, and to identify unsuitable bids. filtering and transferring to the explanation module, the role of the logged-in user to identify and tailor the description and interface the user will see to this role. personalized offers, low-scoring offers, or offers that do not comply with the rules to provide the user with an explanation of cause and effect relationships, 20 explanations based on the technical level the user can understand to simplify or elaborate on, the same explanations offering it in different languages, supporting international users. to provide the generated outputs to the user in a visual and interactive way presenting, by extracting meaning from the user's interactions on the interface 25 system performance is analyzed to generate feedback and the results received Becomes more effective, personalized and understandable over time through feedback. the server (4) configured to enable it to provide explanations a characterized system (1). 9 2. A device such as a smartphone, tablet computer, or laptop computer. A system like the one in Claim 1, characterized by an electronic device (2) (1).

3. Connect to the server (4) using any remote communication protocol. to establish and exchange data with the server (4) through this established connection 5 electronic device (2) configured to make it happen A system like the one in Claim 1 or 2, characterized (1).

4. Connecting to the server (4) via a data network such as the Internet. Claim 3, 10, characterized by an electronic device (2) configured for this purpose. such a system (1).

5. User proposal information, specification criteria, evaluation a system that allows the entry of parameters and user-related operations to present the user interface, the results of the generated proposal evaluation, 15 reasoning behind the decision, natural language customized according to user role. explanations and multilingual explanation outputs are presented to the user graphically and an electronic device configured to provide textual presentation as in any of the above claims characterized by (2). a system (1). 20 6. To be in communication with Server (4) and to be managed by Server (4) The above is characterized by the database structured as (3) a system like any of the requests (1).

7. Offer price information, delivery time information, technical suitability data, specification clauses, bid evaluation scores, user role information, user authorization levels, evaluation results, reasoning behind decisions, natural language description texts, role-based description templates, multilingual descriptions content, user satisfaction data, comprehensibility assessments, 30 user feedback, descriptive learning parameters, artificial intelligence model parameters, description, production history, system operation logs, user transaction logs, system log records and system Data configured to keep a record of configuration parameters any of the above claims characterized by base (3) such a system (1). 5 8. Using any remote communication protocol, with the electronic device (2) to communicate and through this communication with an electronic device (2) with the server (4) configured to perform data exchange a 10 as in any of the above characterized claims system (1).

9. Communication with an electronic device (2) over a data network such as the Internet In Claim 8, characterized by the server (4) configured to establish such a system (1). 15 10. Offer submitted by the user via electronic device (2) server configured to receive evaluation request (4) as in any of the above characterized claims system (1). 20 11. To initiate the bid evaluation process, we will collect bid data and prices. information, delivery time information, technical compliance data, specifications criteria and user role information database (3), cloud server or 25 Ensuring that corporate information is retrieved from an external server (D) The above is characterized by the server (4) configured to do so. a system like any of the requests (1).

12. The received bid data is evaluated according to the specified evaluation rules and machinery. 30 structured to analyze using learning algorithms 11 any of the above requests characterized by the server (4) a system like one of them (1).

13. Calculating bid scores and verifying bids' compliance with specifications. to conduct the evaluation and the resulting 5 server (4) configured to determine the reasons for decision as in any of the above characterized claims system (1).

14. Identify the user's role and, according to the determined user role, 10 the level of detail and technical content density of the statement to be created, characterized by the server (4) configured to determine the form of expression. a system like any of the above-mentioned requests (1).

15. The user can be in different roles such as supplier, manager, purchasing specialist or auditor. Identifying which user role it belongs to and creating The level of detail, technical content, and expression of the statement are as follows: Characterized by the server (4) configured to determine according to the role a system like any of the above requests (1).

16. Proposal evaluation results, scores, specification discrepancies, and decision. to create natural language explanations by using their justifications from the above requests characterized by the configured server (4) a system like any other (1).

17. Generating explanatory texts using natural language generation methods, and then creating the generated texts. explanations, through a role-based language customization mechanism simplify, elaborate, or change the technical level and special ontology-based terminology matching or dictionaries instead of language libraries 30 with server (4) configured to use base jargon filtering 12 as in any of the above characterized claims system (1).

18. Explanations of the large language model sequence-to-sequence learning model, Hybrid 5, where transformer-based models or algorithms are used together. The above characterized by the server (4) to produce with structures. a system like any of the requests (1).

19. The generated descriptions can be translated using machine translation methods or multilingualism. Converting to different languages ​​using description templates and international 10 to enable users to receive explanations in their own languages from the above requests characterized by the configured server (4) a system like any other (1).

20. User satisfaction reported via electronic device (2) 15 their evaluations, their comprehensibility scores, and with server (4) configured to receive textual feedback as in any of the above characterized claims system (1).

21. Analyzing feedback using machine learning algorithms, The explanation is to update the parameters of the production engine and the explanation. with server (4) configured to improve production quality as in any of the above characterized claims system (1). 25 22. The results of the proposal evaluation, the reasons for the decision, and the explanation. texts, user feedback, description generation model parameters, learning outcomes and system transaction logs database (3) storing and traceability, continuous learning and disclosure 30 to ensure that production performance is improved over time 13 from the above requests characterized by the configured server (4) a system like any other (1). 10 20 30

Images