Personalized translation and translation modeling method and system.

Abstract

The invention relates to the steps involved in simultaneous and consecutive translation processes, and to a multi-step process for a personalized translation model, as well as a method and system for personalized translation and interpretation modeling that includes a server, interface, and database. The invention specifically relates to a modeling method and system that, with the assistance of artificial intelligence, enables simultaneous translation process steps and the preparation of a terminology suggestion list through personalized translation modeling between the server and the individual.

Description

1 TARIFF PERSONALIZED TRANSLATION AND INTERPRETATION MODELING METHOD AND SYSTEM The technical field to which the invention relates: The invention is called KÖÇM (Personalized Translation Modeling) (translation from written language to language). translation types (transfer) and under the name of KÖÇM (oral transfer from language to language 5 (the types of translation that exist) and especially (the pressure it creates, time constraints, the translator's Due to its cognitive load and the need for specialized equipment and preparation, it is considered the most difficult to accept. (processes related to simultaneous and consecutive translation) a person who enables its systematic advancement; includes server, interface and database It relates to specialized translation and translation modeling methods and systems. 10 The invention, in particular, addresses the most challenging and critical aspects of translation, both in terms of process and output, with the help of artificial intelligence. a systematic analysis of the source of simultaneous translation problems, which are the type that can create them. software that attempts to prevent this and improves the quality of translation output through personalized analyses. The aim is to ensure that the procurement steps are carried out. The same systematic process can be easily integrated into other oral and written translations. It is created in this type. The Personalized Translation Modeling (PTM) system is artificial. audiovisual data using emotion analysis via neural networks and deep learning capable of analyzing (the gestures, facial expressions, and body language data of the individual being translated) (and also able to read the individual's language in all its nuances within a contextual dimension and target to create the most efficient discursive construction in the language and for translators and interpreters 20 with software that can facilitate and generate the most accurate terminology suggestions possible It is related. State of the art: Personalized translation (oral) and translation (written) modeling methods, translation and interpreting. Preliminary analysis of the individual discourse and cultural elements of the services and 25 in the target culture by being rebuilt (through the practice of discursive engineering) It aims to make it personalized and effective. The main purpose of these methods is... by adapting translation and interpreting processes to the specific needs and preferences of individuals The aim is to achieve a higher degree of accuracy. In translations, this modeling method is used. The speaker's individual language elements include tone of voice, speaking speed, and the 30 most frequently used words. 2 words and phrases, manner of speaking and use of jargon, and all other types of texts he produced. The analysis takes into account characteristics such as type of references and the use of foreign words, etc. This analysis is being conducted to select the type of discourse in the target language that best matches the data derived from this analysis. construction is underway. For example, a speaker's individual language at an international conference. To accurately reflect the style and source culture, the translator must have 5 knowing speech habits, and understanding the characteristics that shape these speech habits and character. It needs to address a number of socio-cultural elements, which in turn affects the audience. It enables them to understand the content of the speech more clearly and effectively. Personalized modeling in written translation involves tailoring texts to the specific language of individuals or organizations. and involves adapting the texts according to stylistic preferences. This process involves 10 to be adapted to the target audience and the institutions requesting the translation. Translation according to a specific language standard, taking into account the translation requests of the individuals. For example, translations done for a brand convey the brand's motivation and It is designed to reflect its style, so that the brand identity and message are consistent. It remains. In addition, the terminology used in personalized translation processes is sector-specific. 15 Grammatical accuracy and terminological consistency are also of great importance, which is crucial for technical processes. applicable across a wide range of areas, from documents to marketing materials. is happening. Technological advancements are further enabling the application of personalized modeling methods. This facilitates it. Especially artificial intelligence and machine learning algorithms, language 20 By analyzing personal preferences and habits in its use, we translate this information into Turkish. It provides a great advantage in terms of application in processes. For example, artificial intelligence. Supported translation tools take into account the user's past translation data and language preferences. By taking these technologies, we can offer more accurate and personalized translations. It enables translators and interpreters to work more efficiently and 25 It improves the quality of personalized services. In conclusion, personalized translation and translation modeling methods enhance translation services. a more accurate and effective approach that better aligns with today's needs a systematic translation used to facilitate communication and developed specifically for each individual. Translation involves preparation, analysis, and the most effective discourse production and transmission strategies in the target language. 30 These methods offer solutions tailored to the language needs of individuals and organizations, thus providing greater... This approach ensures that satisfactory results are achieved in language services. 3 By improving its quality and efficiency, it forms the cornerstone of effective communication in a globalized world. It constitutes. Personalized translation (oral) and translation (written) modeling in the current state of the art. Although various suggestions and applications have been developed for these methods, these improvements are not sufficient. This is not the case. Some applications for inventions developed for this purpose are listed below. 5 is provided. Patent number “WO2017206861A1” exists in the known state of the art. the application, in which the instructions can be repeated and reconfirmed if necessary. It describes the structure. Character input and processing software is added, input and display. Hardware is included and artificial intelligence integration is provided. 10 Instructions can be executed hierarchically, meaning there is a main control and an auxiliary control. This facilitates the separation and allows instructions to be executed more effectively. They are recognizable. People can be recognized, the ability to recognize can be developed, and 'acquaintances' can be identified. It can be stored permanently. Thanks to artificial intelligence, it is more flexible. It is working to ensure that the instructions are executed in a user-friendly manner and that 'acquaintances' are given priority. 15 It is known as a system where 'instructions from acquaintances are given priority'. It includes. Patent application number “CN108255804A”, which exists in the known state of the art, By analyzing the emotional factors that enter natural language, we can target specific emotional responses. It provides production. The system includes a language exchange server, a language communication terminal, and 20 a language structure analysis module containing various databases and a language sentiment analysis module It consists of these modules, which analyze the input information and the language generation module. It sends these materials, and based on these analyses, appropriate language structures and emotional tones are determined. In this way, effective emotional conversations are conducted. By processing language skills, a system and method that provides accurate and targeted responses based on emotional factors 25 It includes. Patent application number “CN110852115A”, which exists in the known state of the art, a simultaneous translation system that integrates artificial intelligence and real-world translation and by offering translation services in a way that suits users' needs. to provide, effectively allocate translator resources, and improve user experience. to develop and support the healthy growth of the online translation market 4 The system aims to include a user terminal, a server, and an interpreter terminal. It includes user terminal, communication, display, sound, translation and feedback. It houses the following modules: server, communication, data storage, machine translation, translator. It includes scoring, order fulfillment, and body learning modules. Translator The terminal itself has communication, display, voice, and translation modules. These modules are 5 a system that enables interaction between the user, server, and interpreter terminals. It includes the system and method. In the current state of the art, a systematic translation and interpretation preparation model and It includes simultaneous and consecutive translation processes with modeling and emotion. Using analysis as well, innovative and comprehensive artificial intelligence and NLP (Natural Language Processing) 10 in the field of text fragments or words micro-macro semiotic nuances By focusing on identifying difficult-to-translate words or phrases, the most Personalized interpretation (oral) and translation including suggestions for suitable sentences or words. A (written) modeling method and system are needed. In conclusion, due to the negative aspects described above and the current solutions, topic 15 Due to its shortcomings, an improvement is needed in the relevant technical field. It has been made. The purpose of the invention: The main aim of the invention is to provide a comprehensive solution for all types of translation and interpreting, especially the most challenging ones. The type of interpreting with types; Conference Interpreting and its two modes (simultaneous and consecutive 20 (translation) artificial intelligence (AI) with a specially created, layered software structure and Simultaneous and consecutive translation processes using machine learning-based NLP. Steps to prepare a CV, create a profile map, analyze individual language, and suggest terminology. personalized translation and interpretation model method and system with list processes It provides. 25 Another aim of the invention is to fill theoretical and practical gaps in translation studies, thereby providing high-quality translation services. It offers translation experience. In this way, it provides insights into translation theories and practices. by filling in the gaps, for both source text producers and translators It aims to provide a high-quality translation experience. Another aim of the invention is to take into account personal linguistic elements and individual discourse characteristics. By doing so, it offers personalized services in simultaneous and consecutive interpreting processes. In this context, simultaneous consideration of personal and semiotic (semiotic) data. Improved performance in interpreting (ET) and consecutive interpreting (AT) processes. is targeted. 5 Another aim of the invention is to reduce the cognitive load and stress of the translator, thereby providing a more accurate representation of the truth. This enables the translator to improve timely translation performance. In this way, the translator... to be able to convey semantic and discursive meaning to the target audience more effectively It provides. Another aim of the invention is to minimize the difficulties encountered in the translation process and 10 It helps improve translation quality by analyzing individual language characteristics. This includes: It provides translator-specific terminology lists and analysis support via the server. and through an interface using artificial intelligence-powered (AI) machine learning. The displayed prepared KÖTM CV and the extracted KÖTM Map are in the individual's language. Analysis and the creation of a terminology suggestion list are ensured. 15 Another aim of the invention is to create an innovative and very comprehensive AI-powered machine learning (NLP). performing sentiment analysis on text excerpts or words in the field It provides. It is not only aimed at negative-positive, neutral emotions, but also at their true essence. Focusing on the micro and macro nuances that trigger its formation - through implication and the body. 20 the ability to use, the overall mood, tone, and manner of speaking that permeates one's character. uncovering the relevant observable behaviors and the conclusions drawn from them by producing the most appropriate contextual response through discursive construction and transferring it to the target language. The process involves creating a list of suggested terminology. Another aim of the invention is to perform two types of functions: coarse-grained mode and macro-router. 25 It is the act of seeing. Regardless of the mode of translation activity; simultaneous or consecutive, the purpose of the speech or the reason for the activity, and the proposal for this invention The concept of interpretum acts as the driving force of the translation process; translation itself, that is... In written translation, the scope of the translation, that is, the reason for its creation, is revealed. The main motivation is to reach the target translation text (translatum). 30 6 Explanation of the figures: FIGURE-1; This drawing shows a schematic representation of the system that is the subject of the invention. FIGURE-2; This diagram shows the flowchart of the Invention Analysis Methodology (IAAM) process. Reference numbers: 100: Interface 5 200: Server 300: Database 400: AI Machine Learning, Deep Learning, Sentiment Analysis, Artificial Neural Networks Creating a KÖTM-KÖÇM corpus using 400.1: Writing the name of KMU 10 400.2: Select the year using the year selection button located in the sub-tabs under this heading via the interface. Selecting the year, year range, or all time by clicking 400.3: Click the Create KÖTM Corpus or Create KÖÇM Corpus button 400.4: Through a server, YZD machine learning retrieves the selected year for the speaker. According to the comprehensive research of historical video data and 15 of this data from the internet withdrawal 400.5: Transcript of KMU's speeches and videos captured via the interface. display of texts as 500: AI Machine Learning, Deep Learning, Sentiment Analysis, Artificial Neural Networks Preparing a CV using KÖTM-KÖÇM 20 500.1: Interface for viewing the E-KÖTM-KÖÇM CV Preparation page. By clicking the "Prepare E-KÖTM-KÖÇM CV" button via 500.2: Viewing the E-KÖTM-KÖÇM CV Preparation page 7 500.3: Selected by clicking on the "Personal Information" text on the page. Page 500.4: Self-discourse, rhetoric, idiolectal language, character, and body language. The psychosocial background that shapes its elements can be selected by clicking on the text. 500.5: On the page, under the heading "Recent Events Attended or Presented" Selected by clicking 5 500.6: Select the number of pages available on the interface for creating the CV. 500.7: Machine learning to create both written and audio versions of KÖTM-KÖÇM CVs. To prepare your CV, click the "CV Preparation" button. 500.8: Machine learning to create both written and audio versions of the KÖTM-KÖÇM CV. preparation as 10 500.9: Viewing the prepared CV via the interface. 500.10: The displayed KÖTM-KÖÇM CV is added to the archive with the person's name. recording 600: Generating a KÖTM-KÖÇM map using machine learning with YZD 600.1: To generate the KÖTM-KÖÇM map again for the speaker via the interface, 15 writing the name 600.2: Select the year from the sub-tabs under this heading via the interface. Selecting the year, year range, or all time by clicking 600.3: Through a server, YZD machine learning retrieves the selected year for the speaker. According to the comprehensive history and psychosocial text and video data research 20 To do this, click the KÖTM button. 600.4: Through a server, YZD machine learning retrieves the selected year for the speaker. According to the comprehensive history and psychosocial text and video data research to be done 600.5: Machine learning to retrieve researched data via a server 25 8 600.6: Collecting all captured data using machine learning. 600.7: Creation of the KÖTM corpus using machine learning. 600.8: Displaying the generated KÖTM (Knowledge, Environment, and Technology) record in the interface. 600.9: To view the Create KÖÇM Corpus page via the interface. Clicking the "Create KÖÇM Corpus" button is step 5. 600.10: Viewing the Create KÖÇM Corpus page via the interface 600.11: Machine learning will collect all written and audiovisual data related to the text to be translated. To start the process, click the "Create KÖÇM Corpus" button. 600.12: All written and audiovisual data related to the text to be translated by machine learning. withdrawal 10 600.13: Creating a Corporate Social Responsibility Corpus using machine learning 600.14: Viewing the generated KÖÇM (Knowledge, Environment, and Culture) Corpus via the interface. 600.15: Machine learning for analyzing audiovisual and written texts using KÖTM-KÖÇM To perform this action and create a CV map, click the analysis button. 600.16: Creation of the KÖTM-KÖÇM map using machine learning 15 600.17: Displaying the KÖTM-KÖÇM map on the interface. 600.18: Saving the displayed KÖTM-KÖÇM map to the database. 700: AI-powered machine learning for individual language analysis and proposition of KÖTM-KÖÇM. preparation of the list 700.1: The KÖTM CV and KÖTM map data found in the database are processed by YZD machine 20 analysis through learning 700.2: AI uses machine learning to generate and translate Source Discourse (SD) in real time. words, phrases, and expressions that are difficult to translate or require prior research determination 9 700.3: Determining the total number of uses within the YZD Corpus Capacity. 700.4: Defining the main context heading used with AI. 700.5: Determining environmental and temporal information using AI. 700.6: Determining the discursive function of the source discourse within the context of general speech. 700.7: YZD machine learning with the most appropriate equivalents in the desired target language 5 Creating a list of suggested terminology 700.8: Displaying the generated list of suggestions on the interface. 800: Creation of Additional Information Text for KÖTM-KÖÇM Using Machine Learning with YZD 800.1: Creating Additional Information Text for KÖTM-KÖÇM via the interface page To view it, click the Create Additional Information Text for KÖTM-KÖÇM button. 10 800.2: Creating Additional Information Text for KÖTM-KÖÇM via the interface page display 800.3: Entering the required additional information requests on the displayed page. 800.4: Additional information requests, if deemed necessary, will be processed by the server. To learn about and withdraw additional information, click the "Create Additional Information Text" button. 15 800.5: Machine learning, extracted information, and manual input by the translator or interpreter. creating additional information text 800.6: Displaying additional information text via the interface 800.7: Saving the displayed Additional Information Text to the database. 800.8: Creating a KÖTM file by printing the output and combining it with the data from the next step. 20 Description of the invention: The invention, with the help of AI, provides a very customized translation model for ET and AT processing steps. explaining all the step-by-step processes and server (200), interface (100) and database (300) It relates to personalized translation and translation modeling methods and systems. The server (200), interface (100) and database (300) are in communication, interface (100) 5 Creating a CV for KÖTM using AI and machine learning-supported NLP methods. This is the person who prepared and created the KÖTM map. Interface (100), running on an electronic device; AI and machine learning Prepared via server (200) with supported NLP, KÖTM-KÖÇM CV and The generated KÖTM-KÖÇM map is displayed and individual language analysis is performed on 10 This element allows the prepared KÖTM-KÖÇM proposal list to be displayed. Database (300), AI and machine-assisted NLP interface (100) The displayed, KÖTM-KÖÇM CV prepared via the server (200) and extracted This is the element where the KÖTM-KÖÇM map is recorded. The invention is particularly and primarily considered the most difficult type of translation, with a result of 15 seconds. A type of interpreting that presents the most challenges for translators due to the necessity of achieving the required results. the execution of AI-assisted processing steps to improve quality in ET. and a terminology suggestion list along with a translation model tailored to the specific needs of the client. It is related to the modeling method that involves preparation. The most difficult type of translation is interpreting by word of mouth. (translation) concept and ET and 20 in real-time and minute-by-minute progress in conference interpreting. Starting with AT and addressing the problems of these types, preparation-process-output and Reducing the cognitive burden on translators creates a more comfortable and efficient translation process. The KÖTM system, which begins with producing solutions in this regard, encompasses all translation (oral translation) minor modifications to the types of translation; secondarily to all types of translation (written translation). It can be implemented by doing this. 25 KÖTM offers a new approach to the translation process. This model is based on KMU (Source). Text Producers determine the purpose of a person's speech, their biographical information, and their primary culture. It involves a detailed analysis. KÖTM (Knowledge, Skills, and Means) involves the translator understanding the individual needs of these people. terminological choices, cultural references, and audiovisual data related to the language (requires prior examination of semiotic elements). This preliminary analysis 30 11 The process allows the translator to better prepare for their work, rather than working alone. and making pre-performance preparations, which were not structured in a certain system, much more systematic. By doing so, it saves time, reduces cognitive load, and reduces stress. It improves translation quality by using KÖTM-KÖÇM File data without being influenced. The interpreter, KMU, directly influences the speech character through idiolectical, semiotic, 5 When you have a comprehensive knowledge of their cultural and individual characteristics, translation They become motivated during this time, and it positively affects their performance. Because with all this prior knowledge, he will systematically implement it during the performance And this will enhance prediction activation, a concept specifically developed for this invention. YZD systematic translation and translation preparation system also provides written or audio output 10 several layered and complex specialized software systems can be acquired sequentially and simultaneously. It includes analysis and processing; for example, deep learning, sentiment analysis, audiovisual. data analysis, text mining, statistics, etc. The system is used by the KÖTM File. studying and synthesizing the data, and then translating or interpreting the data obtained from AI accordingly. An AI translator or 15 that uses AI to improve quality during and after translation. Not just an interpreter; a human being is a professional translator or interpreter. In this context... an AI that is managed and organized by humans as a human-machine hybrid collaboration It is a system that can be considered. AI software can collect big data and large corpus data. he did, but chose words and phrases that caused problems or were the most difficult to translate, or by synthesizing the corresponding expressions or expression patterns and taking initiative in important roles 20 It is a hybrid system where a human translator or interpreter is involved in the task. Personalized Translation Modeling offers solutions tailored to individual and practical needs. By offering these services, it provides translation services that are personalized and have a unique systematic approach. Personalized Translation (PTT) is applicable to all types and activities of written translation. It is also called Modeling (KÖÇM). This model, especially for important KMUs, is 25 By focusing on cultural and rhetorical elements in its translation, it addresses the unique characteristics of each individual. It takes into account the discourse and communication style. KÖTM, VIP speaker (speeches) Wide sphere of influence, global recognition, and international crises due to mistranslations. (which is capable of revealing) the discursive transmission of these important figures in advance By identifying and properly grounding them, the translation process achieves a high 30% success rate. It aims to provide quality and impact. 12 Thanks to software that can provide many helpful functions, translators and interpreters This system can be monitored both before and during the execution of tasks. analysis results and outputs that they can obtain from the screen, or from the outputs, or as an audio recording. Using their reports, we are currently using the latest version of AI software, CHATGPT 4. In situations where 'it' stumbles (in bidirectional translation between low-source languages ​​and especially in 5 This is problematic in ET and AT, which operate on the principle of immediacy. For example, it is commonly used... due to its usage, primarily English, but also German, French, Spanish, etc. structurally (in terms of affix usage, etc.) different from more common and larger source languages and in less commonly used, low-source languages; Arabic-English, Persian-English, Urdu-English, Chinese-English, Turkish-English have a number of structural and semantic differences. This continues to cause problems, especially in the specific regions where a language is spoken. in word and expression usage, cultural nuances, proverbs, idioms, and other aspects of that language In the transfer of expressions and word usages from other languages, different types of texts and In quotations and references made from individuals, each individual requires a personalized preliminary analysis. (in the use of language, discursive elements and rhetorical expressions, etc.) more accurate, efficient and reliable 15 By providing outputs, the translator or interpreter experiences less stress and has a much more comfortable working environment. This process enables powerful prediction activation and is more contextually efficient. It can produce outputs. It can also analyze and process the translation or interpreting tasks performed. The results can be optionally left recorded in the system software, the same applies to KMU. In the new role, which will be in a different environment and subject, the old knowledge will serve as a guide. 20 It is available for use. The KÖTM Analysis consists of a three-stage process and is audiovisual. It includes semiotic (semiotic) and discursive (idiolect) analyses. The stages provide a detailed overview of KMU's linguistic features, cultural context, and communication style. It ensures that it is handled in this way. The model is specific to the translation process and 25 It refers to systematic preliminary preparation and aims to improve the quality of the translation output. It offers various strategies. This process involves the individual language and discursive elements of KMU. to be analyzed within a predetermined framework and the translator's use of this information This ensures the integrity of the navy. In this way, the translation process is efficient and harmonious. is being carried out. In addition, optionally, the Additional Information Text Creation for KÖTM-KÖÇM 30 The procedure is also carried out if deemed necessary. 13 The speaker or team will receive interpreting services from a KÖTM (Knowledge, Skills, and Environment) Translator / Interpreter. Private and corporate clients, such as authors or publishing houses, will be required to provide pre-translation or translation services. related to the topic of the conference event or the context of the text to be translated, or the speech including the full text (text included in translations ET - text included AT or text-free ET - text-free) AT) If deemed necessary, regarding the text or document to be translated in writing, the requested 5 Additional individual or corporate requests regarding KÖTM-KÖÇM services, including translation. Optional additional information about the environment or rules, e.g., participant list, participant professional group, etc., or other determined and needed besides these. This is an additional information text containing lengthy or brief details on the relevant issues. This important addition... The information text includes 10 discursive and behavioral aspects of KMU regarding itself and the subjects it studies. It is possible to support this with all kinds of audiovisual data related to its style. In other words, KÖTM In addition to the written text added to this section of the file, short videos must also be included. In this case, images of KMU, especially those from KT events, must be included. This should be studied. Or, preferably, the audiovisual data provided by KMU. In order to obtain reliable results for the KÖTM Analysis by examining the previous 15 Links to the event videos can also be added to this supplementary information text. After the process step of individual language analysis and preparation of terminology suggestion list (430) Optionally, the Additional Information Text for KÖTM-KÖÇM can be developed using machine learning via YZD. The creation process can be performed. KÖTM-KÖÇM Analysis method process steps include AI machine learning 20 It includes methodological processes consisting of the following steps, using the following methodology: 0- AI machine learning, deep learning, sentiment analysis, artificial neural networks Creating the KÖTM-KÖÇM Corpus using (400), 1- AI machine learning, deep learning, sentiment analysis, artificial neural networks Preparation of KÖTM-KÖÇM CV using (500), 25 2- Creating a KÖTM-KÖÇM map using YZD machine learning (600), 3- AI-powered machine learning for individual language analysis and proposition of KÖTM-KÖÇM (Knowledge, Technology, and Culture, Arts, and Culture) Preparation of list (700), 4- Creation of Additional Information Text for KÖTM-KÖÇM Using Machine Learning with AI (800) (optional output if required). 30 14  The operations to be performed on the relevant page in the KÖTM-KÖÇM Software will be done by a translator and There are two main tabs that translators can navigate to. Written translations For translations, that is, KÖÇM; for oral transmissions, that is, KÖTM or the KÖÇM tab is selected. 0. AI machine learning, deep learning, sentiment analysis, artificial neural networks 5 The creation of the KÖTM-KÖÇM Corpus using;  In the selected tab, first the source text to be translated or translated. Producer's (KMU) / Source Speaker's (KK) Corpus This must be created. To do this, first, the translation must be entered into the browser section of the system. For that, the name of KMU is written. This system is like Google, internet 10 a server that encompasses and integrates into the entire environment (200) It was produced under the name (KÖTM-KÖÇM). It will be translated. When the name of KMÜ is written to the server (200) special KÖTM browser, it appears on the side You can also select details about the years in the boxes as sub-tabs. This can be done. For example, all times, time range – two date entries, 15 Since then, only speech transcripts, related news, related videos, sub-tab boxes such as related books, related scientific articles, related theses Data retrieval related to corpus creation by selecting according to the interpretation. The process is carried out.  If the process involves written translation, the system for the KÖÇM Corpus is 20. The name of the text to be translated is entered into the browser, and if the original version exists... All additional information, including previous authors, is used to create the text for corpus purposes. The name is entered, and in the additional options tab next to it, you can find all the times. time period – two date entries, since …., all versions of the work, relevant news, related videos, related books, translation strategies, related scientific 25 articles, related theses or, if the work is being translated for the first time, the work itself and All information regarding the author is uploaded to the system as a PDF. The work is being published for the first time. If it is to be translated, from the KÖÇM server (200) general or supporting The aim is to create a comprehensive overview of the Turkish Cultural Heritage Foundation by conducting detailed nuance analyses.  Then, the system with interface (100) was 30 with AI machine learning. Transcripts of the relevant KMU speeches from the internet. (His written texts) and existing videos should also be included in the KÖTM Compendium. This is achieved by creating or generating a KÖÇM (Knowledge, Environment, and Compendium) Chart. The KÖTM-KÖÇM Analysis method process steps include AI machine learning, deep Learning, sentiment analysis, artificial neural networks used in the KÖTM-KÖÇM corpus The creation process involves the following steps: 5  Writing the name of KMU (400.1),  Through the interface, click the year selection button located in the sub-tabs under this heading. Selecting the year, year range, or all time by clicking (400.2),  Clicking the Create KÖTM Corpus or Create KÖÇM Corpus button (400.3), 10  Through a server, AI machine learning determines the selected year for the speaker. According to the comprehensive research of historical video data and this data from the internet. withdrawal (400.4),  The written versions of KMU's speeches and videos captured via the interface. Display of texts as (400.5). 15 These are the basic steps: Create Chronicle of KÖTM or Create Chronicle of KÖÇM tabs This is the first step in the 4+1 step analysis to be conducted and it encourages the foundation of it. Therefore, this process should be considered as step 0 or the basic step, and the entire analysis And it should be remembered that the outputs will be obtained from the data extracted from this corpus. 1. YZD Machine Learning: Process of Preparing a CV for KÖTM-KÖÇM (20) steps,  Interface for viewing the E-KÖTM-KÖÇM CV Preparation page By clicking the "Prepare E-KÖTM-KÖÇM CV" button via (500.1),  Viewing the E-KÖTM-KÖÇM CV Preparation page (500.2), 25  Selected by clicking on the "Personal Information" text on the page. (500.3),  The self-expression, rhetoric, and idiolect character found on the page, and On the psychosocial background that shapes body language elements Selected by clicking (500.4), 30 16  On the page, under the heading "Recent Events Attended or Presented" Selected by clicking (500.5),  Select the number of pages available on the interface for creating your CV. (500.6),  Machine learning enables the creation of both written and 5-part CVs for the KÖTM-KÖÇM program. To prepare your CV in a listenable format, click the "Prepare CV" button. (500.7),  Machine learning to create both written and digital versions of the KÖTM-KÖÇM CV. preparation in an audible format (500.8) (with audio and slides if desired) (can also be added as a video), 10  Viewing the prepared CV via the interface (500.9),  The displayed KÖTM-KÖÇM CVs are added to the archive with the person's name. recording (500.10). 2. Generating a KÖTM-KÖÇM map using YZD machine learning.  Entering the speaker's name via the interface, 15  Selecting the year, year range, or all time through the interface,  Through this server, AI machine learning will be used to analyze the speaker's data. Extensive research of historical video data based on the selected year and This data is being extracted from the internet; videos are being filmed for the KÖTM Map. This is done using the key. 20  Next, in the second sub-tab, for the KÖTM map, see Semiotics. The "Analyze discourse data" button is pressed. In this sub-tab, a deeper analysis of discourse data is conducted. The body from the KÖTM software that can perform image and sound analysis through learning. language and its elements, simultaneously the speech character elements of KMU the individual's language elements, discourse, rhetoric, tone of voice, frequency 25 the words, idiomatic expressions and proverbs he uses, the different ways he uses them he produces texts of this type (poetry, anecdotes, couplet readings, aphorisms, etc.). References to various types of texts, all of which are well-known and popular, the discourse, the terminology and expressions chosen in which context, and this includes body language and elements such as gestures and facial expressions. 30 by examining the source according to the principles mentioned in the KÖTM map. It is instructed to report discourse elements. 17  In this way, body language analysts and behavioral science experts The analysis that would take hours to complete can be done in a much shorter time. how the source speaker constructs a self-discourse and how this discourse by focusing on what data it presents that can be observed when it is being created, How KMU and the interpretum will shape and construct their target discourse 5 Information can be obtained that it will be used in this. Finally, this In the tab, two data sets are combined to produce the KÖTM map data. is being removed.  In addition to this mixed information obtained in written form, the following software can be added: It will also provide guidance by creating a similar map, and if desired, the other 10 The information is saved like data in tabs and can be archived later. It can be left for later use. KÖTM-KÖÇM Analysis method processes, specifically the AI ​​machine learning step. To generate the KÖTM-KÖÇM Map using the following steps: includes; 15  To generate the KÖTM-KÖÇM map again for the speaker via the interface writing of the name (600.1),  Through the interface, click the year selection button located in the sub-tabs under this heading. Selecting the year, year range, or all time by clicking (600.2),  Through a server, AI machine learning retrieves 20 data points from the speaker's selected year. According to the comprehensive history and psychosocial text and video data research To do this, click the KÖTM button (600.3),  Through a server, AI machine learning determines the selected year for the speaker. According to the comprehensive history and psychosocial text and video data research to be done (600.4), 25  Extracting researched data through a server using machine learning (600.5),  All data collected was gathered using machine learning (600.6),  Creation of the KÖTM corpus using machine learning (600.7),  Displaying the generated KÖTM corpus on the interface (600.8), 30  To view the Create a Corporate Social Responsibility Chart page via the interface Clicking the Create KÖÇM Corpus button (600.9), 18  Viewing the Create KÖÇM Corpus page via the interface (600.10),  Machine learning will gather all written and audiovisual data related to the text to be translated. To withdraw, click the Create KÖÇM Corpus button (600.11),  Machine learning will process all written and audiovisual data related to the text to be translated. withdrawal (600.12),  Creation of a Corporate Social Responsibility Testing System (CSR) Corpus using machine learning (600.13),  Viewing the generated KÖÇM Corpus via the interface (600.14),  Machine learning for the analysis of audiovisual and written texts using KÖTM-KÖÇM To perform the analysis and create the CV map, click the analysis button. 10 (600.15),  Creation of the KÖTM-KÖÇM map using machine learning (600.16),  Displaying the KÖTM-KÖÇM map on the interface (600.17),  Saving the displayed KÖTM-KÖÇM map to the database (600.18), Which principles of discursive transfer are used in translation activities on the KÖTM-KÖÇM map? 15 Criteria are presented regarding how this can be realized within this framework: A. SOURCE DISCOURSE  Source Text Generator (STG), (Macro that Generates Fine-Grained Modes and Codes) Semiotic Subject - The entity that produces the source discourse (SD) in translation. The source speaker is the primary speaker (SS). In translation, the primary speaker is the first person to write or produce the written text. He is the first author.  KMU - Fine-Grained Modes / Codes Generated by the Macro-Semiotic Subject or Micro Modes / Codes: - In translation, the tone (tension) of the spoken word; in interpretation, the tone of the text. - In translation, the speaker's facial expressions, gestures, and body language (fine 25) granular / micro semiotic modes and codes), the approach taken by the author in translation style, - Tone of speech in translation (fine-grained / micro-semiotic modes) (and codes), and in translation, the way the author describes the work he created, - Speaker's CV (YZD software, previous stage 30) extracted, and here are the micro and macro data 19 (It is also used in synthesis.) In translation, the source text is used. This is the KÖÇM CV of the author who produced the text.  The speech transcripts must be translated using all available electronic resources on the internet. The large-volume corpus of information that makes up the CAT (Common Data Processing) system consists of coarse-grained modes and macroscopic components. It is considered a guiding principle. In translation, however, it refers to a different version of the previous 5 works. The versions, if any, are Coarse-grained mode and Macro Router.  The source medium is Coarse Mode and Macro for both transfer types. It is a guiding factor. It is the environment in which the translation takes place; in translation, it is the text being translated. It is the setting in which it is described or takes place; if these are absent, it is the genre to which it belongs.  Source audience, coarse-grained mode and Macro-Director in translation source 10 The country and people to which the speech, or in translation the original text, belongs.  Source culture expectations, coarse-grained mode and Macro-guide. In translation, it is from the original text produced by the speaker; in interpretation, it is from the written text. the expectations of the country and its citizens from the source text of this type It expresses. 15 B. DISCOURSE MOTIVE (INTERPRETUM)  These are concepts specific to this invention: Interpretum is the interpretive activity of this invention. In the section for, it is used for both translation modes, ET and AT. a translation that is parallel to the reason for producing the speech text to be translated. its motive, that is, the purpose and goal of transmission and the target discourse acquired through this means 20 (HS) output. Translation motivation for written sources to be translated or Its purpose was named skopos by the translation scholar Vermeer. C. TARGET DISCOURSE  Target Discourse Generator (TDG), text analysis into the target language using the KTMM-KÖÇM methodology. They are the people who convey the information by carrying out the stages. This person is KÖTM 25 in translation. For translators or those who do written translations, the term "KÖTM Translator" is used instead of "YZD". Professionals who implement or use EDM software are called experts in their field. (When the YZD KÖTM-KÖÇM user is not a translator or interpreter) Problems in the quality capacity of the targeted translation and translation output. This is because AI is used in the analysis and output generation stages. a person who can combine and synthesize it with their own professional expertise (A professional is needed.) HSU - Macro Semiotic Subject The Fine-Grained Mods / Codes or Micro Mods / Codes it produces are listed below: The titles are expected to be created as tabs in the software:  HSU's production tonnage (Fine-grained particles reflected / detectable in HSU production 5 Modes and Codes), the target constructed by the translator or interpreter in the target language. It is the tone of the text and discourse. Information extracted from the system, and videos if any. Visuals created for translations of the speaker's previous speeches. This will be visible through auditory analysis. Here, the translator or interpreter from the general tone, specific elements and nuances of the spoken or written text 10 or parallel or modifying tones based on speech tension or Mixed tone application where the two can be used in different places depending on the situation. It has the initiative. In this context, in the first and long-term stage, KÖTM- In exceptional cases specific to language and culture, the KÖÇM YZD software may be used. Personal differences are such that even sentiment analysis algorithms are insufficient for personal 15 the power to take initiative in ensuring that differences are reflected in translation And its capabilities will still lag behind the human brain. Therefore, this tone a ton of production that will cause changes in the target language in the software regarding its difference. Change tab or tabs options translate or interpret 20 It will not be used because it may require it. Because before and after execution. You are already a KÖTM Translator or KÖÇM without receiving or listening to the output. The translator synthesizes the data obtained from the software by taking the necessary initiatives. will use a multi-dimensional filter in organizing the relevant nuances. by taking the necessary initiatives, it will prepare for the execution process and 25 It will perform its function. Here, all kinds of AI software are available. Regardless of its level of development, it is multifaceted when compared to a real human being. The initiative (including CHAT GPT 4) cannot be taken, it is personalized and situation-specific. The argument is that they cannot provide guidance. These other cultural or compelling words such as proverbs, idioms, etc. or 30 In cases where it is a cluster: From the previous or current KÖTM Corpus (a List of recommendations derived from the preliminary analysis conducted for the previous event. It can be extracted and synthesized with other new information and used. It's something that's developing at that moment. in this case, an utterance or phrase that is not studied in the corpus or list of propositions 21 (such as a quote from a poem, a proverb, a culturally distinctive expression; a joke, aphorism, etc.) If used, a word-for-word translation can be done at the discretion of the human translator. Or if it would be confusing in that context – just explaining it would create confusion. tons can be produced.  Target environment (MO) (Coarse-Grained Mode and Macro Router), Conference 5 translation and the production of the source oral text in both modes, that is, the source It is the same environment in which the conversation takes place. Because the transmission takes seconds or so. It happens in the same environment within minutes. But written translation in their texts, the translator may choose the environment in which the text is produced, perhaps years later. or immediately after printing, a translation in a different and preferred medium 10 will do so. From the environment in the information and videos, if any, extracted from the system. This will be visible through audiovisual analysis of the translations.  HO guidelines apply to translations of the speech text and video. This can be understood, although not explicitly mentioned, from a detailed examination of the video. 15 that can be understood; where the speech and translation are performed simultaneously or consecutively. These are the rules and principles that define the framework of all types of conference settings. These include the time, place, duration, audience, and other aspects of the speech and translation performance. These are predetermined rules that the speaker is expected to follow. Also... the speaker and their delegation or participants, and the translation performance. The translator, according to his / her interpretation, determined the target discourse construction and 20 the transmission, the practice of the listeners and viewers who will receive the translation It describes the expected rules, etc. (if any). In these written translations... According to the request of private or legal entities receiving translation services from the translator all kinds of rules will be determined according to the translation score that will be formed. It includes. 25  Target audience (Coarse-Grained Mode and Macro-Oriented), translation in translation representing the citizens of the target culture who will receive, listen to or watch In written translation, the target language uses the language into which the translation is made. It represents its readers.  Target global tone (TGT), interpretum in translation, translation efficiency and 30 in their contexts, however, with a purpose framed according to the translation scopus. It is the tone of discourse in the target language being created. To prevent misunderstandings. To avoid causing a political crisis, citing sources (in parallel with KS) in a parallel tone. produced, but without subtracting or adding (explanation as needed). 22 (by taking the initiative or shifting to a mixed tone) a target discourse (TS) tone to create. These framed criteria ensure the speaker is properly understood in the target language. This will ensure that the translation or interpretation is done with an accurate discursive foundation and construction. It is possible. These are the 5 most frequently used elements among the basic individual language elements. words and expressions, of all kinds of cultural vocabulary and phrases (proverbs, idioms, local expressions) (words and expressions), the use of words and expressions from different linguistic origins, poetry, poet including instances where intertextual types such as etc. are introduced; general and specific. the tone of speech, rhetoric, and discourse that vary according to the situation and context, and this It is created by analyzing the subcomponents of phenomena. This data is used for deep learning, 10 Text mining involves sentiment analysis and identifying the speaker's unique speech characteristics. analyzing this software from different angles with other semiotic data specific to this software a layered operational step containing complex algorithms the system processes and accordingly proposes a discourse transfer in the target language YZD offers this software, especially in the speaker's speech or 15 The translator's role is to convey the content of the text to be translated and to construct its discursive structure in the target language. and the words and phrases of all kinds that pose the greatest challenge for translators. The components are found and transferred using a custom model. In other words... The text to be translated or interpreted will be a full, word-for-word translation in AI-powered tools. No, it's about detecting problematic particles or clusters of particles and targeting them in the target language 20 The aim is discursive construction.  In conclusion, this important tab contains the relevant speaker or translation of the AI. from semiotic discourse analysis performed on the person or source to be analyzed The information gathered will be correlated with contextual data to reach a conclusion. 3. Individual language analysis using AI machine learning and KÖTM-KÖÇM 25 Preparation of a list of suggested terminology, 1. In the first sub-tab of this tab, the KÖTM-KÖÇM map from the AI ​​software is available. by considering all the information in the output data - analysis by including all the words most frequently used by the source speaker or KMU and KÖTM-KÖÇM terminology 30, which enables the creation of a list of expression patterns. The "Create suggestion list" button is pressed. After the list appears, the second sub-tab is opened. 23 The HS translation tab allows you to select the source language first, then the target language, and then the interpreter. from the settings, KMU or source speaker's request or main text Depending on the subject matter, translation or a translation motive is added, or if applicable, the relevant It is selected from the tab. After these settings are made, translate or interpret. Pressing the button will generate the KÖTM-KÖÇM terminology suggestion list. 5 If desired, it can be printed out or saved as in the previous tabs. It can also be archived. To summarize, these steps are other important. We can also express it in detail as follows:  The YZD of the KÖTM CV and KÖTM map data found in the database Analysis through machine learning (700.1), 10  YZD generates and translates Source Discourse (SD) using machine learning. words that are difficult to translate immediately or require prior research, Determination of phrase and expression types (700.2),  Determining the total number of uses within the YZD Corpus Capacity (700.3), 15  Determining the main title of the context used with AI (700.4),  Determining environmental and temporal information with AI (700.5),  The discursive function of the source discourse within the context of general speech determination (700.6),  YZD uses machine learning to find the most suitable equivalents in the desired target language. Creation of a terminology suggestion list (700.7)  Displaying the generated list of suggestions on the interface (700.8). 4. Creation of an additional information text for the elective KÖÇM-KÖTM course. Before this, hierarchically, from the beginning forward, the previous step The steps that follow depending on how it is implemented are: 0-1-2-3 25 After these stages, the optional steps for KÖTM-KÖÇM analysis process are as follows: Additional information text can be added to the KÖTM. 2. Selective steps in the KÖTM-KÖÇM Analysis method processes. and the final step is KÖÇM-KÖTM using AI machine learning. If it is deemed necessary to create additional information text, use the tab with the same name in tab 30. It is activated via this. In this tab, again from the previous KÖTM 24 excluding the data obtained from the 4 steps that make up the file content will be provided to the translator or interpreter as an additional information text in the file. The data is available. And like the data in the other stages, it can be used if desired. can be printed out, or electronically if desired. It is in a format that can be stored. 5 3. For this final data sheet, first select the relevant tab, KÖÇM-KÖTM. The "Create information text" button is pressed. In this tab, a human translator or The translator gathered general information from internet research. or simply some other tasks requested of him besides these. (Interpretation of the task, points to consider, task and environment 10) rules, estimated mission duration, performance to be done, and habitat any additional and specific information regarding (number of participants or spectators, (such as slides or text to be used, etc.) and generally Write a text that does not exceed one page and click "create" on the system. It records. 15 4. Thus, the KÖTM is guided by a human translator or interpreter. The analysis system has 4+ optional 1, for a total of 5 steps. It will be completed. 5. This hybrid, human-controlled machine or AI The software mimics neural networks, is capable of deep learning, and has 20 By performing personalized audiovisual analysis, translation and interpretation can be done in many ways. by preventing crises, reducing translator / interpreter stress, and improving the execution process. and a target statement aimed at improving the quality of its output. and is a system of transmission. Specifically, what is called the performance of oral translation. translation and its most difficult form (time, place, importance, instant 25 that does not tolerate errors) Conference interpreting (in terms of processes, stress, and translation output) and In ET and AT, which have two modes, the goal is to prevent crises in a process that operates in seconds. To ensure consistent quality, the translator managing the process is well-prepared. by making it easier to relax, and by simulating predictions in advance a special preparation and process created to increase its activation 30 Personalized translation pre-processing as enhancement software. It is a model. It also involves personalized preparation for the translation process, which is a written process. By offering similar contributions, they can provide similar benefits. The process of creating the Additional Information Text for KÖTM-KÖÇM involves the following steps: includes;  Through the interface, the Create Additional Information Text for KÖTM-KÖÇM page To view, click the "Generate Additional Information Text for KÖTM-KÖÇM" button. clicks (800.1), 5  Through the interface, the Create Additional Information Text for KÖTM-KÖÇM page display (800.2),  Entering the requested additional information on the displayed page (800.3),  The machine will process any additional information requests deemed necessary via the server. To learn about and withdraw additional information, click the "Create Additional Information Text" button. 10 (800.4),  Machine learning, combined with extracted information and manual input by the translator or interpreter. Creating additional information text (800.5),  Displaying additional information text via the interface (800.6),  Saving the displayed Additional Information Text to the database (800.7), 15  Creating a KÖTM file by printing the output and combining it with the data from the next step (800.8). The generated file is in electronic format, just like in all the previous steps. The e-KÖTM file or the printed KÖTM file can be accessed in either format, depending on your needs. It is available for use.

Claims

26 REQUESTS 1. A person who provides personalized translation and translation preparation assistance. It is a specialized translation and interpretation system, its distinguishing feature being;  AI and machine learning performed on an electronic device Prepared via server (200) with supported NLP KÖTM-KÖÇM 5 Its budget, KÖTM-KÖÇM CV, and the issued KÖTM-KÖÇM By performing an individual language analysis, it is possible to visualize their map. This allows viewing of the prepared KÖTM-KÖÇM Recommendation List. interface (100),  Displayed via interface (100) with AI and machine-assisted NLP, 10 KÖTM-KÖÇM CV prepared via server (200) and extracted Database in which the KÖTM-KÖÇM Map is recorded (300),  In communication with the interface (100) and the database (300), the interface (100) NLP deep learning supported by AI and machine learning. Using these methods, create a KÖTM-KÖÇM corpus and submit a CV of 15. a server that prepared and produced the KÖTM-KÖÇM Map (200) It includes.

2. Personalized translation and interpretation modeling incorporating KÖTM analysis processes. It is a method, and its characteristic is;  AI machine learning, deep learning, sentiment analysis, artificial neural network 20 To create a KÖTM-KÖÇM Corpus using networks (400),  AI machine learning, deep learning, sentiment analysis, artificial neural network Preparation of KÖTM-KÖÇM CV using networks (500),  YZD mapping the KÖTM-KÖÇM map using machine learning removal (600), 25  Individual language analysis and KÖTM-KÖÇM using machine learning with AI Preparation of a list of suggestions (700),  Additional Information Text on KÖTM-KÖÇM using machine learning by YZD Its creation involves (800) steps. 27 3. This is a personalized translation and interpretation modeling method in accordance with Claim 2, and its characteristic feature is; AI machine learning, deep learning, sentiment analysis, artificial neural networks Creating the KÖTM-KÖÇM Corpus using (400) process steps;  Writing the name of KMU (400.1),  Select the year from the sub-tabs under this heading via the interface. 5 Select the year, year range, or all time by clicking the button. (400.2),  Click the Create KÖTM Corpus or Create KÖÇM Corpus button clicks (400.3),  Through this server, AI machine learning can retrieve 10 speaker data. Extensive research of historical video data based on the selected year and Extraction of this data from the internet (400.4),  KMU's speeches recorded via the interface and Displaying the text of the videos in written form (400.5) process It includes the following steps. 15 4. This is a personalized translation and translation modeling method in accordance with Claim 2, and its characteristic feature is; AI machine learning, deep learning, sentiment analysis, artificial neural networks In the process step (500) of preparing the KÖTM-KÖÇM CV using;  To view the E-KÖTM-KÖÇM CV Preparation page Click the "Prepare E-KÖTM-KÖÇM CV" button via the interface (20). clicks (500.1),  Viewing the E-KÖTM-KÖÇM CV Preparation page (500.2),  Selected by clicking on the "Personal Information" text on the page. (500.3),  The self-expression, rhetoric, idiolectal language character found on the page and 25 On the psychosocial background that shapes body language elements Selected by clicking (500.4),  The "Recent Events Attended or Presented" section on the page Selected by clicking on it (500.5),  To create a CV, select the number of pages available on the interface: 30 (500.6), 28  Machine learning to create both written and digital versions of the KÖTM-KÖÇM CV. To prepare your CV in a listenable format, click the "Prepare CV" button. clicks (500.7),  Machine learning to create both written and digital versions of the KÖTM-KÖÇM CV. preparation in a restable state (500.8), 5  Viewing the prepared CV via the interface (500.9),  The displayed KÖTM-KÖÇM CVs are archived in the database by name. The process involves adding and saving (500.10) the steps involved.

5. This is a personalized translation and translation modeling method in accordance with Claim 2, and its characteristic feature is; Mapping of KÖTM-KÖÇM using YZD machine learning (600) 10 in the process step;  The speaker can revisit the KÖTM-KÖÇM map via the interface. writing his name for removal (600.1),  Select the year from the sub-tabs under this heading via the interface. Selecting the year, year range, or all time by clicking the button 15 (600.2),  Through a server, AI machine learning provides information to the speaker. Comprehensive history and psychosocial texts and videos, categorized by the selected year. To conduct a data analysis, click the KÖTM button. (600.3), 20  Through a server, AI machine learning provides information to the speaker. Comprehensive history and psychosocial texts and videos, categorized by the selected year. research of the data (600.4),  Machine learning of data researched through a server withdrawal (600.5), 25  All data collected was gathered using machine learning (600.6),  Creation of the KÖTM corpus using machine learning (600.7),  Displaying the generated KÖTM corpus on the interface (600.8),  Create a Corporate Social Responsibility Chart page via the interface To view it, click the Create KÖÇM Corpus button. 30 (600.9), 29  Create a Corporate Social Responsibility Chart page via the interface display (600.10),  Machine learning will analyze all written and audiovisual data related to the text to be translated. To retrieve the data, click the "Create KÖÇM Corpus" button. (600.11), 5  Machine learning will analyze all written and audiovisual data related to the text to be translated. data retrieval (600.12),  Creation of a Corporate Social Responsibility Testing System (CSR) Corpus using machine learning (600.13),  Viewing the generated KÖÇM (Knowledge, Environment, and Culture) Corpus via the interface (600.14), 10  Machine learning in audiovisual and written texts analysis for conducting the analysis and creating a CV map Clicking the button (600.15),  Creating a KÖTM-KÖÇM map using machine learning (600.16), 15  Displaying the KÖTM-KÖÇM map on the interface (600.17),  Saving the displayed KÖTM-KÖÇM map to the database. (600.18) includes the steps of the process.

6. This is a personalized translation and translation modeling method in accordance with Claim 2, and its characteristic feature is; Individual language analysis and KÖTM proposition list 20 using machine learning. in the preparation (700) process step;  The YZD of the KÖTM CV and KÖTM map data found in the database analysis through machine learning (700.1),  AI creates Source Discourse (SD) using machine learning and Immediate translation is difficult or requires prior research for translation. 25 Determining the types of words, phrases and expressions that require (700.2),  Total number of uses within the YZD Corpus Capacity determination (700.3),  Determining the main title of the context used with AI (700.4),  Determining environmental and temporal information with AI (700.5), 30  Determining the discursive function of KS within the general context of speech. (700.6),  YZD uses machine learning to find the most suitable option in the desired target language. Creating a list of terminology suggestions with their equivalents (700.7),  Displaying the generated list of suggestions on the interface (700.8) process It includes the steps.

7. This is a personalized translation and translation modeling method in accordance with Claim 2, and its characteristic is; 5 In the process step (800) of Creating Additional Information Text for KÖTM-KÖÇM;  Through the interface, the Create Additional Information Text for KÖTM-KÖÇM page To view, click the "Generate Additional Information Text for KÖTM-KÖÇM" button. clicks (800.1),  Through the interface, page 10 of the Creating Additional Information Text for KÖTM-KÖÇM display (800.2),  Entering the requested additional information on the displayed page (800.3),  Requests for additional information may be made via the server if deemed necessary. Creating Additional Information Text for Machine Learning-Based Data Extraction Clicking the button (800.4), 15  Machine learning, the extracted information and the translator's or interpreter's Creating additional information text by manually entering it (800.5),  Displaying additional information text via the interface (800.6),  Saving the displayed Additional Information Text to the database (800.7),  The output is collected and combined with the data from the next step to create the KÖTM file 20 The creation (800.8) process includes steps.

8. This is a personalized translation and translation modeling method in accordance with Claim 7, and its characteristic feature is; The process involves creating a KÖTM file by printing out the data and combining it with the data from the next step. In this step, as in all previous steps of the process, electronically Both 25 files can be selected as e-KÖTM files or as output KÖTM files, depending on your needs. It includes the possibility of using it in this way.

Images