The Intelligent System Development for Psychological Analysis of the Person's Condition Oksana Oborskaa, Vasyl Andrunyka, Liliya Chyruna, Ruslan Haskoa, Anatolii Vysotskyib, Solomiia Mushastaa, Oksana Petruchenkoc and Iryna Shakleinad a Lviv Polytechnic National University, S. Bandera Street, 12, Lviv, 79013, Ukraine b Anat Company, Chervona Kalyna Avenue, 104, Lviv, 79049, Ukraine c Hetman Petro Sahaidachnyi National Army Academy, Heroes of Maidan Street, 32, Lviv, 79012, Ukraine d Drohobych Ivan Franko State Pedagogical University, Ivan Franko Street, 24,Drohobych, 82100, Ukraine Abstract A system for the individual psychological and emotional state analysis is developed. The aim is to assess the individual through practical and recommendations social networks. The assessment data, this area problems and the system's relevance analysis are studied. The diagrams are developed that describe the system logic and structure. System requirements and a prototype application description are done According to RUP methodology, which stimulates an individual analysis system activity is created. Keywords 1 Social network, information system, intelligent system, information technology, psychological analysis, personality trait, emotional state, big data, psychological state, system analysis, information resource, text analysis, data processing, social network personal analysis, Facebook profile, psychological type, modern machine learning technology, psychological portrait, decision making 1. Introduction Life poses professional and personal problems to the person constantly. It is not enough to have the skills, the mind, the gift, and the honesty to succeed. Success also depends on a person's self-esteem, attractiveness, competence, competitiveness, communication skills, ability to be a team player. In market relations, for a modern specialist, a psychological culture, the main components of knowledge of oneself, another person, a culture of behaviour and communication, is no less important than, for example, knowledge of a personal computer or knowledge of a person a foreign language. The success of affairs depends on the psychological culture of human. It is relevant in today's world in the age of social networking, where accounts have become an inexhaustible source of information about each individual. Posts on Twitter, Vkontakte, and Facebook posts, even subscriptions to certain bands and music, are valuable information that allows you to find out a nearly complete psychological portrait of a person before getting to know her. Social network users often behave differently than in real life, which is one of the problems in determining psychological and emotional state. On the other hand, given this fact, you can understand how a person sees himself in the outside world and what points of interaction with him can found. Therefore, it becomes urgent to learn how to behave COLINS-2021: 5th International Conference on Computational Linguistics and Intelligent Systems, April 22–23, 2021, Kharkiv, Ukraine EMAIL: oksana.v.oborska@lpnu.ua (O. Oborska); vasyl.a.andrunyk@lpnu.ua (V. Andrunyk); lchirun21@gmail.com (L. Chyrun); ruslan.v.hasko@lpnu.ua (R. Hasko); anat1957@gmail.com (A. Vysotskyi); solomiyanytrebych@gmail.com (S. Mushasta); voksanietko@gmail.com (O. Petruchenko); ioshakleina@gmail.com (I. Shakleina) ORCID: 0000-0001-9606-0267 (O. Oborska); 0000-0003-0697-7384 (V. Andrunyk); 0000-0003-4040-7588 (L. Chyrun); 0000-0001-9280- 8098 (R. Hasko); 0000-0001-9190-7051 (A. Vysotskyi); 0000-0003-4932-4113 (S. Mushasta); 0000-0003-2304-8149 (O. Petruchenko); 0000-0003-0809-1480 (I. Shakleina) ©️ 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (CEUR-WS.org) in a social network to feel as if they are a socially successful person. It is necessary to learn how to use social networks properly and learn to filter public content because the ability to present this information psychologically correctly largely depends on the success of each person. There are few information resources and programs available to help determine a person's psychological or emotional state through social networks. Mostly, it comes down to apps or programs that offer some tests to determine your character, "compatibility" with your friends, and more. However, they all have one major flaw - they are not automated. Everyone is required to undergo tests each time, and often the results of these tests are different. A person strives to minimise their efforts to obtain the information they seek. It is more convenient for the user, by installing the application or by registering on the information resource, to be constantly acquainted with assessing their psychological or emotional state. Therefore, the construction of information systems that will perform this work is relevant. The purpose of the work is to create a system for analysing the personal user data (messages on social networks, tweets, etc.) and building a psychological portrait of a person with the generation of conclusions and recommendations of the psychological and emotional state of the individual. 2. Related works Social networks are becoming more and more popular. A considerable number of people have accounts in several of them at once. With the help of social networks, people find their friends, socialise, share interests, share information. At the same time, actively using social networks, a person informs himself of various content. Is it possible to say anything specific about the person based on this information or to use it, for example, for job evaluation? The idea of evaluating a person's personality by their activity in social networks is increasingly capturing the minds of researchers. Several such attempts have recently made. Donald Kluemper and his companions published an article [1] in 2012describing how three specially trained evaluators more or less successfully identified the personal traits of Facebook users on their profile. The researchers set out to test whether the owner of the Big Five characteristics could be determined based on an analysis of a Facebook profile, predict the likelihood of hiring that person for work, and their future effectiveness, although the predictive validation of expert estimates was low and the sample was small. Many unique features are reflected in the Facebook profile, according to the authors. For example, the number of friends may be associated with extraversion, people with a high level of awareness are more likely to be more cautious about the content of messages and comments they write online, and people with expressed kindness will be more trusting, so their profile will contain more information that is personal. To test these assumptions, three specially trained evaluators, each of whom completed two-hour training, were asked to analyse 274 profiles of real people on the Facebook network. In addition, assess based on this information is the holders of these profiles of each Big Five features (neuroticism, extraversion, openness to experience, goodwill, awareness) and determine the likelihood of hiring them for a starting position in the service industry. Six months after the evaluation procedure, the researchers attempted to contact the leaders of the evaluated study participants. They asked them to fill in questionnaires about the effectiveness of the employee and his or her "civic" behaviour toward colleagues and the organisation as a whole. Such data is collected from 56 executives. The results showed that the estimates of personality traits obtained from the experts are significantly related (r = 0.23-0.44) to the estimates obtained by the personal questionnaire. Thus, analysing a Facebook profile is a good way to evaluate the personality traits of its owner. A significant correlation is also found between the personality traits of a Facebook profile and the performance estimates received from executives. The most predictable power is emotional stability (r = 0.27) and goodwill (r = 0.31). Among the estimates of personality traits obtained from the most evaluated, the only extraversion correlates with work efficiency (r = 0.28). Thus, Facebook profile analysis has proven to be a more valid predicting performance than personal questionnaires. Considering that it took an average of 5 to 10 minutes to analyse one profile of assessors, this method of personality assessment is desirable to use in the selection process. However, its use involves some legal and ethical issues. For example, during an interview, the employer may not ask questions regarding race or religion, sexual orientation, marital status, etc. When analysing the same profile on the social network, this information can become obvious and affect the hiring decision. However, researchers at the University of Pennsylvania (https://www.upenn.edu/) and The Psychometrics Centre - University of Cambridge (https://www.psychometrics.cam.ac.uk/) have done. In September 2013, they published an article [2] describing the results of the analysis of 700 million words, phrases and topics collected from Facebook posts by 75,000 people. The analysis showed striking differences in the frequency of different words and phrases between people of different sex, age, and personality traits. Here, for example, the differences between men and women are shown in Figs. 1a, the differences between age groups (13– 18, 19–22, 23–29, 30–65) are shown in Fig 1b. However, the differences between extroverts and introverts, between neurotics and emotionally stable are shown in Figs. 2. This research is part of the World Well-Being Project, in which several studies based on social network analysis have made and are being carried out. a) b) Figure 1: а) The differences between men and women and b) age differences [2] Figure 2: Differences between different emotional types [2] The Psychometrics Centre - University of Cambridge (https://www.psychometrics.cam.ac.uk/) is also trying to predict the individual characteristics of a person by liking it on Facebook. An analysis of such preferences of 58,000 people showed [3] that their model assumes belonging to:  White Americans or African-Americans in 95% of cases,  Gender - in 93% of cases,  Sexual orientation - in 88% of cases in men and 75% of subjects in women,  Democrats or Republicans in 85% of cases,  Christians or Muslims in 82% of cases. The accuracy of the prediction of the remaining dichotomous variables shows in Fig. 3a. In terms of personality traits, the predictive validity is lower: intelligence (r = 0.39), extraversion (r = 0.40), openness (r = 0.43). However, it is quite comparable to the validity of personality tests. The accuracy of the prediction of other individual features shows in Fig. 3b. These results suggest that many tasks for assessing certain specific characteristics of a person (including personality traits) can analyse their profile and activity on social networks. In doing so, it will not be a person but a computer. In recruitment, for example, every second person searches for a candidate's social networking profile before inviting him for an interview. Of course, there are difficulties and limitations. First, as soon as the majority learns that and with what accuracy their social media profiles can be determined, the majority will either cease using them actively, or create several accounts, or, using social networks, use a specific strategy by creating a certain image of yourself (such is the socially desirable use of Facebook). There are also ethical issues. However, the prospects are astounding. a) b) Figure 3: Prediction of a) individual specifics and b) other characteristics of the user's Facebook [3] Analysis of known remedies. Psychological research is very relevant in the modern world, especially in developing information technologies and social networks. However, there are currently no software or information systems on the software market that would address this comprehensively and thoroughly. All existing applications can \ divided into the following subgroups: 1. WEB applications and online tests. This software functions as a standard test: you are asked to take a test and answer specific questions. Based on these answers, the program finds a match in its database or, using different algorithms, determines the percentage composition of the answers, and selects the most competitive one. The problem with such programs is that they are not complete information systems [4-15]. Often, this is generally a separate web page with a set of scripts. Such programs cannot learn and analyse vast amounts of information [16-20]. 2. Statistical systems. This software can handle large data sets and generate statistics. For example, you could show a sample of the most common words and phrases [16-20]. The person carries out further analysis of the data and, on its basis, makes the appropriate conclusions. There is no complete computer analysis and system training [21-33]. The following resources can be exemplified as appropriate systems:  Istio.com - allows you to get advanced text analysis by word, output the top most used words and other text information. Figure 4: The results of the text's analysis on Istio.com  Info.seocafe.info - resource allows you to define the basic parameters of the text that designed mainly for SEO optimisation (Fig. 5). Figure 5: The results of the text's analysis on Info.seocafe.info. The advantages of such resources are ease of use - the user simply enters the required text, and the system automatically processes it [34-39]. However, the disadvantage is that these systems do not allow you to determine a person's psychological or emotional state because they are intended for statistical processing of the text [40-43]. Based on this classification, the following features of the system should be considered: 1. The system must process large amounts of data [44-48]. 2. There is a need for a mechanism to store the data for further processing [49-50]. 3. The system must be intelligent (i.e. it must be trained based on previous data) [52]. 4. The system must be able to perform statistical data processing [53]. 5. The system should be accessible to the public [54]. 6. The final software should function as standalone software (not an application for a particular social network or browser extension) [55]. 7. The system must draw a clear distinction between social networks (that is, be universal) [1-20]. Given all of the above, it is possible to note the relevance of the work. It will be helpful to ordinary users as well as to employers or other public services. 3. Material and methods The software system functionality. The mechanism of use of functions allows describing the possibilities of this information system (Table 1, Fig. 6). To improve the work with this information, we introduce the concept of function attributes - data elements (Table 2), which will provide additional information about each function (Table 3). Table 1 System functions No Name Description(function value) 1 Personalisation Each program must strictly personalised to the specific person that is each user should create an account to simplify the functioning of the information system 2 Protecting personalised When creating an account, the user will provide information sensitive information protected from unauthorised use by third parties. 3 Primary data analysis Collecting and processing data from a profile on a social network. Key metric filtering 4 Statistical data processing Identifying critical patterns in the collected data and compiling a statistical report 5 Training Based on the collected data and previous experience - the refinement of the program's previous findings or creation of new ones 6 Storing the data that has been Providing access to pre-analysis data for the user processes 7 Character analysis Identifying critical indicators of a person's character 8 Analysis of temperament Identification of critical indicators of a person's temperament 9 Analysis of emotionality Identification of critical indicators of a person's emotionality 10 Analysis of self-esteem Identifying critical indicators of a person's self-esteem 11 Analysis of volitional qualities Determination of critical indicators of a person's volitional qualities 12 Analysis of social qualities Identifying critical indicators of a person's social qualities 13 Comparison and adjusting the Based on features 7-12, compare previous results with results current results and adjust them if necessary 14 Formulation of general The generalisation of the psychological state of the conclusions individual 15 Making recommendations Identifying the necessary strategies for communicating with the individual or other interaction features 16 Reporting a study error Identifying a study error and keeping it informed 17 Help on using the program Creating complete support for the user 18 Correcting possible analysis errors The ability for certain users with the necessary authority to correct possible errors in the mechanisms of psychological analysis or the results 19 Determination of psychological Typing of personality by Sigmund Freud's research and type by S. Freud classification 20 Jung's definition of psychological Typing of personality by research and classification of type Jung 21 Encyclopaedia and information on Creation of the necessary reference book, which psychology contains a description of all the psychological characteristics of the individual 22 Data backup Data backup function for cases where this data can be lost (personal data, research data) 23 The function of accessing data and The ability to interactively compare data from different comparing data from different social networks social networks 24 Personal Analysis The ability to personally analyse a personalised user, both based on their social media accounts and queries 25 Finding people by specific types or Based on specific preferences, search or filter people on preferences social networks Table 2 Attributes of functions No Description Attribute (Function Value) 1 Suggested. Status (determination of the final version of the feature approval) Approved. On. 2 Critical. Priority (importance of function) Important. Useful 3 Low. Average. Difficulty (complexity of function implementation) High. 4 Risk (the likelihood that implementing a function will cause undesirable Low. Average. effects such as cost increases, changes in the implementation schedule, High. etc.) 5 Low. Average. Stability (the likelihood that this function will change over time) High. 6 Target version (the version of the product where the feature first Version number appears) 7 Purpose (comments for developers that improve their understanding of Comment development) Based on the analysis of known means of solving the problem and taking into account the features of IP, it is possible to distinguish the finite functions (Table 3). Figure 6: Node tree Table 3 System functions N Priority Status Purpose Stability Risk Employment Version 1 Critical Included data personalisation High Low Average 1.0 2 Critical Included protection of High Low Average 1.0 personal data 3 Important Included filtration of data High Average Average 1.0 4 Important Included identifying key High Average Average 1.0 patterns 5 Important Included intellectual learning Average High High 2.0 of the information system 6 Important Included access to previous Average Average Average 3.0 program results 7 Critical Included definition of Average High High 1.0 personality 8 Critical Included determination of Average High High 1.0 personality temperament 9 Critical Included definition of Average High High 1.0 emotional personality 10 Critical Included definition of self- Average High High 1.0 esteem of the individual 11 Critical Included determination of Average High High 1.0 volitional qualities of personality 12 Critical Included definition of social Average High High 1.0 qualities of personality 13 Critical Included comparison and Average High High 1.0 correction of results 14 Critical Included general conclusions Average High High 1.0 15 Critical Included recommendations Average High High 1.0 16 Useful Included determination of Average Low Low 3.0 errors in the study 17 Important Included creation of help Low Low Low 2.0 18 Critical Included bug fixes Average High High 3.0 19 Important Proposed classification by S. Average High High 4.0 Freud 20 Important Proposed Jung classification Average High High 4.0 21 Important Proposed encyclopaedia Average Average Average 4.0 22 Critical Proposed data backup Average Average High 4.0 23 Useful Proposed personal analysis Average Average High 5.0 24 Important Proposed interactive data Average Average High 5.0 comparison from different social networks 25 Useful Proposed search and filter Average High High 6.0 or people by a specific more psychological type Considering all of the above, it should be noted that, as there are currently no qualitative and comprehensive solutions on the information technology market that offer a solution to this problem, it is incredibly relevant. The purpose of research in this area is to assess the psychological state of modern society. It is possible to create severe information systems (IPs) that, based on social networks, will determine and predict the so-called "temperature" - the general state of society at a certain point in time. It will help avoid all kinds of problems related to social dissatisfaction of the population, etc., to find your application to solve a wide range of issues and tasks in various fields. 4. Experiments, results and discussion 4.1. System analysis Fig. 7 shows a diagram of the use case diagram. The use case diagram uses two types of basic entities: use cases and actors, between which the following types of relationships are established: association - between the actor and the use case; generalisations between actors; abstraction between use cases; inclusion between use cases. System requirements description according to RUP methodology: 1. Interested persons of the precedent and their requirements: a. Project Administration: wants complete information on IP functioning. b. A person whose psychological analysis is being conducted: does not mind that their data will be analysed (i.e., it gives open access to their account); c. Registered user: wants to quickly learn the psychological state of a person without spending a lot of time; 2. The IP user, that is, the main actor in this precedent. This ordinary person is a registered user and performs psychological analysis of different persons through IP. 3. Preconditions: a. The data of the person whose psychological analysis will investigate should be open. b. The user must complete the IP authorisation (or registration) procedure; c. IP must be active; 4. The main successful scenario (Fig. 8): a. IP searches for a person and retrieves information from their account; b. The user enters the name or ID of the person whose social network account is being investigated; c. IP conducts data analysis; d. The user begins a new psychological study; e. IP displays detailed analysis results on the screen. Figure 7: Use case diagram 5. Scenario Extensions or Alternative Streams: a. Invalid person ID I. The IP searches and displays the account of the person found (this is the point of return to the main scenario). II. The user re-enters the correct person identifier. III. If necessary, the user can ask the IP and obtain (as a hint) the complete list of possible person identifiers (for example, those that begin with a specific number or letter). IV. The IP notifies the user of the error and cancels the input of the requested person. b. No person found I. If the person is not found again, the IC notifies the user of the error and returns to the initial state. II. If necessary, the user can ask the IP and obtain (as a hint) the complete list of possible person identifiers (for example, those that begin with a specific number or letter). III. The IP notifies the user of the error and cancels the input of the requested person. c. Person restricted access to their data I. The IP queries the search for another person or returns to its original state. II. The IP notifies the user of an access error. Figure 8: Activity diagram 6. Post-conditions: this is a list of conditions that must always be met in the case of successful execution of the main scenario (i.e., when the interests of all interested parties under item 2 are satisfied), for example: a. The session has successfully recorded in the IP database. b. The user has completed this session. c. The necessary recommendations and conclusions are presented. d. Data on psychological analysis of personality are processed and saved in IP. 7. Special scenarios: a. Provide special access to a privileged group of users to correct malfunctions. b. Ensure that all sessions are as reliable as possible. c. Ensure that the IP user interface is localised. d. Provide 100% data storage capability. 8. List of necessary technologies and additional devices: a. IP must be developed for all existing desktop and mobile operating systems. b. IP must be submitted as an application for all existing browsers. c. IP should be designed as a WEB-oriented system. A sequence diagram describes a process of psychological analysis of a person initiated by a particular user of the system. The user logs in and queries the person's search database. The IS finds the person, processes the data and returns the result. Before shutting down, the system records this session to its database (Fig. 9). Figure 9: Sequence diagram Fig. 10 shows an example of a commented UML dependency packet diagram that depicts a typical WEB-based IC architecture for working with a database and system decision logic. Figure 10: Package diagram 4.2. Functional point's analysis We distinguish the functions of the software and count the number of factors: 1. External inputs: Three (practical recommendations, methods of psychoanalysis, ways of presenting results). 2. External outputs: Three (psychological portrait, study error information, research-based recommendations). 3. External requests: One (classification of psychological types). 4. Internal logical files: Two (preliminary analysis data, personal account data). 5. External logical files: Three (text, data from other social networks, data on Internet activity). The resulting values are then multiplied by the complexity coefficients for each factor (according to IFPUG) and are summarised. The importance of these coefficients is given in Table. 4. Table 4 The value of the coefficients of complexity Odds Difficult Medium Easy Exterior entrances 6 4 3 External requests 6 4 3 Internal logical files 15 10 7 Exterior entrances 7 5 4 External logical files 10 7 5 Accordingly, the considered example in Table 5 shows the parameter values. Table 5 Example parameter values Easy Medium Difficult Number Coefficient Number Coefficient Number Coefficient Exterior entrances 1 3 1 4 1 6 External requests 1 3 0 4 0 6 Internal logical files 1 7 1 10 0 15 Exterior entrances 1 4 1 5 1 7 External logical files 1 5 1 7 1 10 Function Size: AF = 1 × 3 + 1 × 4 + 1 × 6 + 1 × 4 + 1 × 5 + 1 × 7+ 1 × 3 + 1 × 7 + 1 × 10 + + 1 × 5 + 1 × 7 + 1 × 10 = 71. This number is a preliminary estimate and needs correction by assigning weights (0 to 5) to each project characteristic. In addition to the functional requirements of the product are imposed system- wide requirements that limit developers in the choice of solution and increase the complexity of development. An equalisation factor (VAF) is used to account for this complexity. The VAF factor depends on 14 parameters that determine the system characteristics of the product. These 14 system parameters (degree of influence, DI) are rated on a scale of 0 to 5. For the example considered, these characteristics are given in Table 6. Table 6 Example of characteristic values No Characteristic Value 1 Communication 1 2 Distributed data processing 3 3 Productivity 0 4 Hardware limitations 2 5 Transaction load 3 6 Intensity of user interaction 5 7 Ergonomics (end-user performance) 5 8 Flexibility 3 9 Complexity of processing 4 10 Reuse 4 11 Ease of installation 5 12 Ease of administration 3 13 Need for multiple installations in different conditions 5 14 The intensity of data change (ILF) by users 3 Calculating the total effect of 14 system characteristics (total degree of influenza, TDI) is a simple summation of 46. The formula calculates the specified available size: VAF = AF × (0.65 + 0.01 × TDI) and is as follows: VAF = AF × (0.65 + 0.01 × 29) = 66.74. The resulting VAF value can then be converted to a unit of measure (code line number, SLOC), or a performance factor can be estimated because of FP per day (Performance factor), based on which project complexity can be estimated. 4.3. Work scheduling in the Gantt Project software The Gantt Project software allows you to visually analyse the development of a particular software product or system and properly allocate the resources and time of project staff. The tool avoids unwanted anomalies such as untimely completion of the project, misallocation of forces and more. The primary step is to subdivide the project into sub-tasks and identify their executors (Fig. 11). a) b) Figure 11: a) List of project participants and b) subdivision of the project After all the subtasks are created, we get the diagram shown in Fig. 12. It is also possible to review and check whether the resources in the project were allocated correctly (Fig. 13). Figure 12: Gantt chart Figure 13: Allocation of resources in the project For all the above actions, we obtain a PERT diagram showing the sequence of software creation processes as a result (Fig. 14). Figure 14: PERT diagram Choosing a software design methodology is a critical point in developing IP analysis of a person's psychological state (Fig. 15). The design process determines which ways and ways the system will be implemented. It describes the sequence of work and the distribution of labour. By choosing the correct methodology, you can develop optimal options for creating an information system. This IP will be geared towards Internet applications, so the choice of tools and implementation tools considered the technologies that will allow you to implement the required software product. For a long time, not all Internet technologies had a defined standard. Of course, with the advent of HTML5 and CSS3 and the release of Microsoft's MS Edge browser, the situation is changing dramatically. However, there is still a problem with the backward compatibility of sites. When choosing an architectural solution, it is necessary to use two parts, server and client. Both pieces are closely linked and function as one. However, processing of the primary process will take place in the server part (Fig. 16). Figure 15: Context IDEF0 diagram for the specified process Figure 16: IDEF0 diagrams for primary process subtasks (IDEF0 decomposition) In this case, the server hardware is a remote server. There is no need to create a separate server for the project because the program does not have data processed on the server-side. In the future, its implementation is necessary. The client part will be contained on the user's device (PC, laptop, tablet or another device) and displayed in the browser. When designing a Web site, you should consider how it behaves on different devices, with varying diagonals of the screen and other browsers. Recently, tablets and smartphones have become particularly popular as the most mobile devices. Viewing the content on the therefore, the so-called cross-platform and cross-browser are used to solve these problems. Figure 17: IDEF0 diagrams for main process tasks (Text Analysis block decomposition) Figure 18: Systematic scripts - IDEF3 charts. Cross-platform allowed to adapt the website to screens with small diagonals and focused on the site only the most necessary content of the result of the analysis of the person's psychological state. It reduces the load time of the web page, which is one of the critical points. Cross-platforming can be achieved through media queries and the use of "rubber" layouts. A Cross-browser is designed to make the site appear the same on all popular browsers, as each browser has its technologies for implementing specific functions. Therefore, different individual scripts and requests are used to make the site look the same in any browser. Today, the following are popular: Google Chrome; Mozilla Firefox; Opera; Safari; Internet Explorer; Microsoft Edge. The situation will finally change with the release of Windows 10 and the new Microsoft Edge browser. The new browser promises to get around even the current leader Google Chrome. In addition, it will support all the modern WEB standards. The next step is to choose a software solution. Certainly, hypertext mark-up language (HTML) and cascading style sheets (CSS) will be used when designing any web site. As virtually all browsers have recently started to support the basic standards, it is worth paying attention to versions of these technologies, namely HTML5 and CSS3, which prevent the use of outdated technologies such as Flesh on the site. Writing programs is stopped in JavaScript, which has been popular with web developers in recent years. Therefore, the following technologies are selected to create the page layout (otherwise called front-end): HTML5, CSS3, and JavaScript. To create a back-end, you should use technologies such as PHP programming language and MySQL commands. The implementation of the software is performed using the following three software components: HTML, CSS and JavaScript. HTML is required to create the site structure. CSS is a language that defines a website's visual appearance; JavaScript is a programming language needed to implement what cannot achieved in CSS and provide site interactivity. The main file is the index.html file, which is the site's main page (Fig. 19). All major blocks of the site will placed in so-called containers. They are required to centre the content on a web page (Fig. 20-21). They are described all in the main CSS file. JavaScript must use to create features such as user authorisation and word statistics, as the visual appearance and content of some of the blocks on the site changes during the execution process. This cannot implemented with cascading styles only. In addition, certain values and identifiers are created in the process, which are necessary for the proper functioning of the program. a) b) Figure 19: Web site structure After creating all the necessary features, we get a site template, which, after testing, is ready for use (Fig. 22-23). Fig. 24 shows a diagram of an IC database for the analysis of a person's psychological state. By analogy with the definition of the amount of information, we introduce the following concepts. The level of awareness of the decision-maker (ODA), i.e. the doctor, indicates the expert's level of knowledge about the subject of analysis or research. Figure 20: DF diagrams Figure 21: DF diagrams (decomposition) a) b) Figure 22: a) User authentication and b) File selection process Figure 23: The process of interpreting the results of the analysis. Figure 24: Database. (EER diagram) Quantitatively, the level of awareness of ODA will be characterised by the magnitude of the change in the level of uncertainty of knowledge of obtaining information. Understanding ATS means a difference in the level of uncertainty about the situation or subject of analysis of receiving information as a result. With the receipt of data, the level of uncertainty of the problem may decrease if the notification is accurate but may increase if it is deliberately distorted or untrusted (i.e. not confirmed by experience, calculations, documents or otherwise). In addition, the uncertainty of the situation can be assessed based on the objectives of systematic analysis, particularly in terms of determining the extent and level of risk. For each condition, the following types of uncertainty of knowledge characterise:  associated with the possibility of a situation;  described by the quality of available and new information received;  related to the degree of impact of a problem on the level of risk. It follows that raising the level of awareness of ODA does not always lead to the disclosure of the uncertainty of the occurrence of a particular situation, as it was accepted during the formation of relation (1) [21-33]. Therefore, it is advisable to determine the level of awareness of ODA, considering all the above factors. First, let's define the inverse of the level of understanding. Formula (1) establishes an entire group of random events or random states [21-33]. It is taken into account that entropy is the inverse of the amount of information. The value of H is a measure of the uncertainty of a set consisting of n random events with probabilities p1,…,pn. Formula (1) follows that H = 0 provided that only one event will occur from the set of circumstances, and other events can't coincide [21-33]. Such condition is fulfilled in the consecutive transmission of the message in letters [21-30]. 𝐻 = − ∑𝑛𝑖=1 𝑝𝑖 𝑙𝑜𝑔 𝑝𝑖 . (1) We analyse the impact of the quality of information on the level of awareness of ODA. It should be noted that the assessment of the quality of data is the least studied in both computer science and other disciplines, one way or another related to information: the theory of optimal management, the idea of decision making, and so on. Today there is no accepted system of indicators for assessing the quality characteristics of information. Therefore, it is not advisable to dwell on the analysis of the various approaches to formalising them since they do not apply to the solution of most practical system analysis problems. Here are only the essential qualitative properties of information crucial to system analysis tasks, particularly for assessing the degree and level of risk in common, freelance, and critical situations. The definitions in Table 7 should be taken into account when defining ODA awareness indicators [21- 33]. Table 7 The main qualitative properties of information for the psychological analysis of a person No Name Definition 1 Uncertainty a property that reflects the presence of several alternative descriptions of the situation 2 Inaccuracy a property indicating that there is a specific interval of tolerances or errors in measurements or calculations in the quantitative parameters and/or qualitative characteristics of the situation description 3 Incomplete a property that reflects the presence of information gaps in the description of the situation (something missed, not described enough, etc.) 4 Blurred a property that characterises the vagueness of describing a situation where it is impossible to accurately determine the presence or absence of a particular property or its exact quantitative characteristic (for example, it is unbelievable to quantify concepts such as comfortable weather, favourable situation accurately - their description is subjective, vague) 5 Timeliness property that characterises the relationship in time between the moment of occurrence of an event and the moment of receipt of information about it; If ODA does not have enough time to form and make decisions based on the information received, then it is untimely 6 Unreliability a property that reflects the presence of quantitative data or qualitative characteristics that do not correspond to the actual state of the situation 7 Controversy a property that indicates the presence of quantitative or qualitative characteristics that have meaning or content that contradicts other data Let's analyse some techniques and the essence of the uncertainties of the occurrence of situations. We will assume that the level of non-awareness is the uncertainty of knowledge about the emergence of an alternative from the predicted set of problems. The anticipation of expertise can be estimated based on different approaches. Let the set of possible situations Ms is discrete, and each element Si of the set Ms characterises a certain probability pi for 𝑖 = 1, 𝑚𝑠 Then the value of non-awareness Hs will be defined as the level of uncertainty of information about Ms. Therefore, we have a condition similar to that for formula (1) [21-30]. Therefore, uncertainty can be defined as entropy 𝑚𝑠 (2) 𝐻𝑆 = − ∑ 𝑝𝑖 𝑙𝑜𝑔 𝑝𝑖 𝑖=1 Note that for equally probable events 𝑝𝑖 = 1/𝑚𝑆 and 𝐻𝑆 = 𝑙𝑜𝑔 𝑚𝑆 [21-30]. In the process of functioning of the system 𝐹 = {𝐹𝑗 |𝑗 = 1, 𝑚} under the influence of many uncontrolled risk factors 𝐹𝑗 , the staff situation 𝑆𝑖 can turn into a critical, emergency or catastrophic [21]. Such a transition may occur over some time, the duration of which is unknown at a priori and which depends on the number, properties and duration of the influence of the factors 𝐹𝑗 ∈ 𝐹. It is necessary to determine such a permissible period 𝑇0 for the formation and implementation of a solution for which the probability of a situation 𝑆𝑖 transition to a critical, emergency or catastrophic one will not exceed the set value 𝜂 = 𝜂𝑎𝑑𝑑 . The number of risk factors and situations will be given in Table 8, where the "+" sign means that when the relevant factor influences the staff situation becomes critical, emergency or catastrophic, and the "-" sign does not affect the case. Note that the method and algorithm for solving the problem are applicable to finite values i and j. The likelihood of a situation 𝑆𝑖 transitioning under the influence of a factor 𝐹𝑗 ∈ 𝐹; 𝑗 ∈ [1; 7] in a 𝑖𝑗 critical, emergency or catastrophic situation depends on the change in the timing of completeness 𝐼𝑃 , reliability 𝐼𝐷 and timeliness 𝐼𝑇 of ODA awareness. The probability ij of such an event determines the 𝑖𝑗 𝑖𝑗 ratio [21]: 𝜂𝑖𝑗 = 1 − 𝑙𝑔[1 + 𝛼𝑖𝑗 𝐼𝑖𝑗 (𝑡)]; 𝐼𝑖𝑗 (𝑡) = 𝐼𝑃𝑖𝑗 (𝑡)𝐼𝑇𝑖𝑗 (𝑡)𝐼𝐷𝑖𝑗 (𝑡). Table 8 Risk factors that affect the transition of a staff situation to a critical or catastrophic one Errors in the appropriate Absence of timeframe Test failed guidelines literature Database specialist Incorrect Incorrect relevant Фj practice Lack of failure text Si Text Analysis + + + + + + - Compare and adjust + + + - - - - results Primary and secondary processing + - - - + - + of results Forming conclusions and + - - + - - + recommendations 𝑖𝑗 𝑖𝑗 𝑖𝑗 To make a decision, it is necessary to find a rational compromise between levels 𝐼𝑃 , 𝐼𝐷 , 𝐼𝑇 to shorten 𝑖𝑗 the time for its formation and implementation [21]. The indicators of completeness 𝐼𝑃 and reliability 𝐼𝐷𝑖𝑗 of ODA awareness increase over time and are defined by the following conditions [21]: 𝑖𝑗 𝑖𝑗 𝑖𝑗 𝐼̂𝑃 (1 + 𝛼𝑖𝑗 𝑡), 𝑖𝑓 0 < 𝐼̂𝑃 (1 + 𝛼𝑖𝑗 𝑡) < 1, 𝐼𝑃 (𝑡) = { 1, 𝑖𝑓 𝐼̂𝑃𝑖𝑗 (1 + 𝛼𝑖𝑗 𝑡) ≥ 1; 𝑖𝑗 𝐼̂𝐷𝑖𝑗 (1 + 𝛾𝑖𝑗 𝑡), 𝑖𝑓 0 < 𝐼̂𝐷𝑖𝑗 (1 + 𝛾𝑖𝑗 𝑡) < 1, 𝐼𝐷 (𝑡) = { 1, 𝑖𝑓 𝐼̂𝐷𝑖𝑗 (1 + 𝛾𝑖𝑗 𝑡) ≥ 1. At the same time as the time of influence of factors 𝐹𝑗 ∈ 𝐹 decreases the level of the indicator of timeliness of awareness 𝐼𝑇𝑖𝑗 according to its properties which characterises the ratio 𝑖𝑗 𝐼̂𝑖𝑗 (1 − 𝛽𝑖𝑗 𝑡 2 ), 𝑖𝑓 0 < 𝛽𝑖𝑗 𝑡 2 < 1, 𝐼𝑇 (𝑡) = { 𝑇 0, 𝑖𝑓 𝛽𝑖𝑗 𝑡 2 ≥ 1. Therefore, the length of time for the formation, adoption and implementation of the ODA decision is reduced to prevent the transition of the investigated situation to a critical, emergency or catastrophic one. The coefficients 𝛼𝑖𝑗 , 𝛽𝑖𝑗 , 𝛾𝑖𝑗 characterise the dynamics of changes in awareness indicators. They are determined by the dependencies [21]: 𝑒 𝛼̑ 𝑖𝑗 𝐼̂𝑃𝑖𝑗 0,5, 𝑖𝑓 0 < 𝛼𝑖𝑗 ≤ 1, 𝛼𝑖𝑗 = { 0, 𝑖𝑓 𝛼𝑖𝑗 > 1; ̂𝑖𝑗 −5 (𝛼̂ + 𝛾𝑖𝑗 )𝐼𝑇 10 , 𝑖𝑓 0 < 𝛼̂𝑖𝑗 + 𝛾𝑖𝑗 ≤ 1, 𝛽𝑖𝑗 = { 𝑖𝑗 0, 𝑖𝑓 𝛼̂𝑖𝑗 > 1. 𝑖𝑗 𝑒 𝐼̂𝐷 𝛼̂𝑖𝑗 0,05, 𝑖𝑓 0 < 𝛼̂𝑖𝑗 ≤ 1, 𝛾𝑖𝑗 = { 0, 𝑖𝑓 𝛼̂𝑖𝑗 > 1. The values 𝐼̂𝑃𝑖𝑗 , 𝐼̂𝐷𝑖𝑗 , 𝐼̂𝑇𝑖𝑗 are preliminary estimates of the relevant indicators, which experts determine at the time of detection of a freelance model of operation of the system [21], and the coefficients 𝛼̂𝑖𝑗 characterise the influence level of each of the factors 𝐹𝑗 ∈ 𝐹; 𝑗 ∈ [1; 7] on the properties of situations 𝑆𝑖 , 𝑖 ∈ [1; 4]. The values of indicators 𝐼̂𝑃𝑖𝑗 , 𝐼̂𝐷𝑖𝑗 , 𝐼̂𝑇𝑖𝑗 and coefficient 𝛼̂𝑖𝑗 are given in Table 9. Table 9 𝒊𝒋 𝒊𝒋 𝒊𝒋 The values of indicators 𝑰̂𝑷 , 𝑰̂𝑫 , 𝑰̂𝑻 and coefficient𝜶 ̂ 𝒊𝒋 Фj 𝐹1 𝐹2 𝐹3 𝐹4 𝐹5 𝐹6 𝐹7 Si 𝛼̂𝑖𝑗 𝑆1 0.40 0.65 0.85 0.60 0.50 0.75 - 𝑆2 0.45 0.70 0.80 - - - - 𝑆3 0.50 - - - 0.60 - 0.60 𝑆4 0.55 - - 0.75 - - 0.60 𝑖𝑗 𝐼̂𝑃 𝑆1 0.50 0.65 0.85 0.65 0.55 0.75 - 𝑆2 0.55 0.70 0.85 - - - - 𝑆3 0.60 - - - 0.65 - 0.70 𝑆4 0.65 - - 0.80 - - 0.70 𝐼̂𝐷𝑖𝑗 𝑆1 0.55 0.70 0.90 0.70 0.60 0.80 - 𝑆2 0.60 0.80 0.85 - - - - 𝑆3 0.65 - - - 0.65 - 0.75 𝑆4 0.70 - - 0.80 - - 0.70 𝑖𝑗 𝐼̂𝑇 𝑆1 0.60 0.80 0.90 0.75 0.65 0.80 - 𝑆2 0.65 0.90 0.90 - - - - 𝑆3 0.70 - - - 0.70 - 0.80 𝑆4 0.75 - - 0.85 - - 0.75 0,375 0,627 1,004 0,596 0,456 0,801 0 0,434 0,711 0,955 0 0 0 0  ij  0,497 0 0 0 0,596 0 0,642 0,567 0 0 0,854 0 0 0,642 0,033 0,063 1,00 0,058 0,044 0,080 0 0,039 0,071 0,095 0 0 0 0  ij  0,046 0 0 0 0,058 0 0,061 0,053 0 0 0,084 0 0 0,061 0,260 0,570 0,855 0,493 0,353 0,664 0 0,318 0,694 0,805 0 0 0 0  ij  10  4  0,382 0 0 0 0,461 0 0,529 0,452 0 0 0,709 0 0 0,496 To determine the length of the allowable period 𝑇0 = [𝑇1 ; 𝑇2 ], where both 𝑇1 and 𝑇2 the lower and the upper bounds of the interval solve the inequality: 0 ≤ 1 − 𝑙𝑔 (1 + 𝛼𝑖𝑗 𝐼𝑇𝑖𝑗 𝐼𝐷𝑖𝑗 𝐼𝑃𝑖𝑗 (1 + 𝛼𝑖𝑗 𝑡)(1 + 𝛾𝑖𝑗 𝑡)(1 − 𝛽𝑖𝑗 𝑡 2 )) ≤ 𝜂𝑎𝑑𝑑 . Table 10 Valid intervals 𝑻𝟎 for decision formation Фj 𝐹1 𝐹2 𝐹 𝐹4 𝐹5 𝐹6 𝐹7 Si 𝑆1 [0; 30.5] [0; 40.5] [0; 46.6] [0; 40.2] [0; 34.5] [0; 40.7] - 𝑆2 [0; 32.6] [0; 44.6] [0; 48.4] - - - - 𝑆3 [0; 35.9] - - - [0; 38.5] - [0; 39.5] 𝑆4 [0; 40.2] - - [0; 42.4] - - [0; 39.9] For the situation, the time allowed for the formation, decision-making and implementation of the decision should not exceed T1=34.5; for S2-T2=32.6; for S3-T3=35.9; for S4-T4=39.9. Software testing is performed modularly. It is a static analysis of the program, which is the construction of a graph (Fig. 25), the elements of which are the parts of the software to be tested. Figure 25: The managing graph Testing of the tool will be carried out in a monolithic way (simultaneous combination of all modules that form the system into one testing complex). In the general case, if the function is not performed, it will be blocked and will need further correction. This testing method is most appropriate since there is no need to break it down into separate complexes. After developing the interface part and testing, a general overview of the resource's performance is provided. When accessing the resource, the user gets to the authorisation page (Fig. 26-27). Here is a general description of the help and the actual authorisation field. If the authorisation fails, the user will see an error. Figure 26: Authorisation on the site After successful authorisation, the user is taken to the resource's home page, where his personal information will be displayed and a block where the user can select the file for analysis and analyse it. To analyse the text, select the text file and click the "Analyse!" Button. Figure 27: Main Page Figure 28: File Upload File analysis in the text, as a result, the ten most used words in the text will be displayed. In addition, based on the comparison of keywords, the person's psycho-emotional state will be deduced. Therefore, the site contains only the necessary information and does not create difficulties for the user when working with it. All functions are clear and straightforward. Figure 29: Output of results and interpretation of results of text analysis 5. Conclusions In the course of the work, an information system is developed, through which it is possible to conduct a psychological analysis of a person using his messages from social networks. The system helps to automate the process of gathering information and obtaining results. The subject area is analysed, various information systems based on a particular principle are considered. A description of the system requirements according to the RUP methodology is given, a detailed description of the main functions of the system, an algorithm of operation, and side effects. The necessary software environment description in which the information system is developed is made and the implementation of the software. The information system meets all modern requirements and allows users to analyse the text and formulate the necessary conclusions. 6. References [1] D. H. Kluemper, P. A. Rosen, K. W. Mossholder, Social Networking Websites, Personality Ratings, and the Organisational Context: More Than Meets the Eye?, volume 42(5) of Journal of Applied Social Psychology, 2012, pp. 1143-1172. [2] H. A. Schwartz, J. C. Eichstaedt, M. L. Kern, Personality, Gender, and Age in the Language of Social Media: The Open-Vocabulary Approach, volume of 8(9) Plos One, 2013. URL: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0073791 [3] M. Kosinski, D. Stillwell, T. Graepel, Private traits and attributes are predictable from digital records of human behavior, volume 110(15) of Proceedings of the National Academy of Sciences, 2013, pp. 5802–5805. https://www.pnas.org/content/110/15/5802 [4] V. Vysotska, L. Chyrun, Analysis features of information resources processing, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2015, pp. 124-128. [5] A. Gozhyj, I. Kalinina, V. Vysotska, V. Gozhyj, The method of web-resources management under conditions of uncertainty based on fuzzy logic, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2018, pp. 343-346. [6] O. Kanishcheva, V. Vysotska, L. Chyrun, A. Gozhyj, Method of Integration and Content Management of the Information Resources Network, volume 689 of Advances in Intelligent Systems and Computing, 2018, pp. 204-216. [7] A. Gozhyj, V. Vysotska, I. Yevseyeva, I. Kalinina, V. Gozhyj, Web Resources Management Method Based on Intelligent Technologies, volume 871 of Advances in Intelligent Systems and Computing, 2019, pp. 206-221. [8] J. Su, A. Sachenko, V. Lytvyn, V. Vysotska, D. Dosyn, Model of Touristic Information Resources Integration According to User Needs, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2018, pp. 113-116. [9] V. Lytvyn, V. Vysotska, Designing architecture of electronic content commerce system, in: Proceedings of the International Conference on Computer Science and Information Technologies, CSIT, 2015, pp. 115-119. [10] V. Vysotska, R. Hasko, V. Kuchkovskiy, Process analysis in electronic content commerce system, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2015, pp. 120-123. [11] V. Vysotska, V.B. Fernandes, M. Emmerich, Web content support method in electronic business systems, volume Vol-2136 of CEUR Workshop Proceedings, 2018, pp. 20-41. [12] J. Su, V. Vysotska, A. Sachenko, V. Lytvyn, Y. Burov, Information resources processing using linguistic analysis of textual content, in: Proceedings of the International Conference on Intelligent Data Acquisition and Advanced Computing Systems Technology and Applications, Romania, 2017, pp. 573-578. [13] V. Vysotska, V. Lytvyn, Y. Burov, A. Gozhyj, S. Makara, The consolidated information web- resource about pharmacy networks in city, volume Vol-2255 of CEUR Workshop Proceedings, 2018, pp. 239-255. [14] B. Rusyn, V. Vysotska, L. Pohreliuk, Model and architecture for virtual library information system, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2018, pp. 37-41. [15] B. Rusyn, V. Lytvyn, V. Vysotska, M. Emmerich, L. Pohreliuk, The Virtual Library System Design and Development, volume 871 of Advances in Intelligent Systems and Computing, 2019, pp. 328-349. [16] L. Chyrun, I. Kis, V. Vysotska, L. Chyrun, Content monitoring method for cut formation of person psychological state in social scoring, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2018, pp. 106-112. [17] V. Vasyliuk, Y. Shyika, T. Shestakevych, Information System of Psycholinguistic Text Analysis, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 178-188. [18] S. Fedushko, M. Davidekova, Analytical service for processing behavioral, psychological and communicative features in the online communication, volume 160 of The International Workshop on Digitalization and Servitization within Factory-Free Economy (D&SwFFE), 2019, pp. 509-514. [19] M. Z. Zgurovsky, N. D. Pankratova, System analysis: Theory and applications. Springer Science & Business Media, 2007. [20] M. Z. Zgurovsky, Y. P. Zaychenko, The fundamentals of computational intelligence: System approach. Springer International Publishing, 2017. [21] M. Z. Zgurovsky, N. D. Pankratova, Systems analysis. Problems, methodology, applications [in Russian], 2010. [22] M. Z. Zgurovsky, V. S. Mel'nik, P. O. Kasyanov, Evolution Inclusions and Variation Inequalities for Earth Data Processing I: Operator Inclusions and Variation Inequalities for Earth Data Processing, volume 24 of Springer Science & Business Media, 2010. [23] M. Z. Zgurovsky, A. D. Gvishiani, K. V. Yefremov, A. M. Pasichny, Integration of the Ukrainian science into the world data system, volume 46(2) of Cybernetics and Systems Analysis, 2010, pp. 211-219. [24] M. Z. Zgurovsky, P. O. Kasyanov, O. V. Kapustyan, J. Valero, N. V. Zadoianchuk, Evolution Inclusions and Variation Inequalities for Earth Data Processing III: Long-Time Behavior of Evolution Inclusions Solutions in Earth Data Analysis, volume 27 of Springer Science & Business Media, 2012. [25] V. A. Sadovnichiy, M. Z. Zgurovsky, Modern Mathematics and Mechanics: Fundamentals, Problems and Challenges. Springer, 2018. [26] M. Z. Zgurovsky, A. A. Pavlov, The four-level model of planning and decision making, in: Combinatorial Optimization Problems in Planning and Decision Making, 2019, pp. 347-406. [27] M. Z. Zgurovsky, V. G. Totsenko, V. V. Tsyganok, Group incomplete paired comparisons with account of expert competence, volume 39(4-5) of Mathematical and Computer Modelling, 2004, pp. 349-361. [28] V. A. Sadovnichiy, M. Z. Zgurovsky, Advances in dynamical systems and control. Springer International Publishing, 2016. [29] V. Lytvynenko, I. Lurie, J. Krejci, M. Voronenko, N. Savina, M. Ali Taif. Two Step Density-Based Object-Inductive Clustering Algorithm. Proceedings of the Second International Workshop on Modern Machine Learning Technologies and Data Science (MoMLeT& DS-2019), Shatsk, Ukraine, June 2-4, 2019, pp. 117-136. ISSN 1613-0073. URL: http://ceur-ws.org/Vol-2386/ [30] I. Lurie, V. Lytvynenko, S.e Olszewski, M. Voronenko, W. Wóicik, O. Boskin, U. Zhunissova, M. Sherstiuk, Application of Inductive Bayesian Hierarchical Clustering Algorithm to Identify Brain Tumors, volume 1246 of Advances in Intelligent Systems and Computing, 2021, pp. 567-584 DOI: 10.1007/978-3-030-54215-3_36 [31] V. Demchenko, S. Olszewski, M. Voronenko, E. Zaets, N. Savina, I. Lurie, V. Lytvynenko, Modeling and predicting the organochlorine pesticides concentration in the child's body based on their accumulation in the mother's body, volume 2631 of CEUR Workshop Proceedings, 2020, pp. 419-432. [32] S. Fedushko, M. Gregus, T. Ustyianovych, Medical card data imputation and patient psychological and behavioral profile construction, in: Proceedings of the 9th International Conference on Current and Future Trends of Information and Communication Technologies in Healthcare (ICTH), 160, 2019, pp. 354-361. [33] V. Lytvyn, V. Vysotska, I. Peleshchak, T. Basyuk, V. Kovalchuk, S. Kubinska, L. Chyrun, B. Rusyn, L. Pohreliuk, T. Salo, Identifying Textual Content Based on Thematic Analysis of Similar Texts in Big Data, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2019, pp. 84-91. [34] V. Danylyk, V. Vysotska, V. Lytvyn, S. Vyshemyrska, I. Lurie, M. Luchkevych, Detecting Items with the Biggest Weight Based on Neural Network and Machine Learning Methods, volume 1158 of Communications in Computer and Information Science, Springer, Cham, 2020, pp. 383-396. DOI: https://doi.org/10.1007/978-3-030-61656-4_26 [35] O. Duda, N. Kunanets, O. Matsiuk, V. Pasichnyk, N. Veretennikova, A. Fedonuyk, V. Yunchyk, Selection of Effective Methods of Big Data Analytical Processing in Information Systems of Smart Cities, volume Vol-2631 of CEUR Workshop Proceedings, 2020, pp. 68-78. [36] A. Lutskiv, N. Popovych, Big Data Approach to Developing Adaptable Corpus Tools, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 374-395. [37] O. Naum, L. Chyrun, O. Kanishcheva, V. Vysotska, Intellectual System Design for Content Formation, in: Proceedings of the International Conference on Computer Sciences and Information Technologies, CSIT, 2017, pp. 131-138. [38] V. Lytvyn, P. Pukach, І. Bobyk, V. Vysotska, The method of formation of the status of personality understanding based on the content analysis, volume 5/2(83) of Eastern-European Journal of Enterprise Technologies, 2016, pp. 4-12. [39] V. Lytvyn, V. Vysotska, V. Shatskykh, I. Kohut, O. Petruchenko, L. Dzyubyk, V. Bobrivetc, V. Panasyuk, S. Sachenko, M. Komar, Design of a recommendation system based on Collaborative Filtering and machine learning considering personal needs of the user, volume 4(2-100) of Eastern- European Journal of Enterprise Technologies, 2019, pp. 6-28. [40] T. Batiuk, V. Vysotska, V. Lytvyn, Intelligent system for socialization by personal interests on the basis of SEO technologies and methods of machine learning, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 1237-1250. [41] N. Shakhovska, S. Fedushko, M. ml. Greguš, N. Melnykova, I. Shvorob, Yu. Syerov, Big Data analysis in development of personalised medical system, in: Proceedings of the 10th International Conference on Emerging Ubiquitous Systems and Pervasive Networks (EUSPN), 160, 2019, pp. 229-234. [42] A. Berko, V. Alieksieiev, V. Lytvyn, Knowledge-based Big Data Cleanup Method, volume Vol- 2386 of CEUR Workshop Proceedings, 2019, pp. 96-106. [43] N. Shakhovska, R. Kaminskyy, E. Zasoba, M. Tsiutsiura, Association rules mining in big data, volume 17(1) of International Journal of Computing, 2018, pp. 25-32. [44] O. Veres, N. Shakhovska, Elements of the formal model big date, in: Proceedings of the International Conference on Perspective Technologies and Methods in MEMS Design, MEMSTECH, 81-83. (2015) [45] N. Shakhovska, The method of big data processing, in: Proceedings of the 12th International Scientific and Technical Conference on Computer Sciences and Information Technologies, CSIT, 2017, pp. 122-126. [46] A. Rzheuskyi, O. Kutyuk, O. Voloshyn, A. Kowalska-Styczen, V. Voloshyn, L. Chyrun, S. Chyrun, D. Peleshko, T. Rak, The Intellectual System Development of Distant Competencies Analysing for IT Recruitment, volume 1080 of Advances in Intelligent Systems and Computing IV, Springer, Cham, 2020, pp. 696-720. [47] O. Lozynska, V. Savchuk, V. Pasichnyk, Individual Sign Translator Component of Tourist Information System, volume 1080 of Advances in Intelligent Systems and Computing IV, Springer Nature Switzerland AG, Springer, Cham, 2020, pp. 593-601. [48] N. Antonyuk, M. Medykovskyy, L. Chyrun, M. Dverii, O. Oborska, M. Krylyshyn, A. Vysotsky, N. Tsiura, O. Naum, Online Tourism System Development for Searching and Planning Trips with User's Requirements, volume 1080 of Advances in Intelligent Systems and Computing IV, Springer Nature Switzerland AG, 2020, pp. 831-863. [49] V. Lytvyn, V. Vysotska, Y. Burov, V. Hryhorovych, Knowledge novelty assessment during the automatic development of ontologies, in: Proceedings of the IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP, 2020, pp. 372-377. [50] Y. Burov, V. Lytvyn, V. Vysotska, I.a Shakleina, The Basic Ontology Development Process Automation Based on Text Resources Analysis, in: Proceedings of the IEEE 15th International Scientific and Technical Conference on Computer Sciences and Information Technologies, CSIT, 2020, pp. 280-284. [51] M. Emmerich, V. Lytvyn, V. Vysotska, V. Basto-Fernandes, V. Lytvynenko, Preface: Modern Machine Learning Technologies and Data Science (MoMLeT+DS 2020), volume Vol-2631 of CEUR Workshop Proceedings, 2020. [52] V. Vysotska, Ukrainian Participles Formation by the Generative Grammars Use, volume Vol-2604 of CEUR workshop proceedings, 2020, pp. 407-427. [53] O. Bisikalo, V. Vysotska, V. Lytvyn, O. Brodyak, S. Vyshemyrska, Y. Rozov, Experimental Investigation of Significant Keywords Search in Ukrainian Content, volume 1293 of Advances in Intelligent Systems and Computing V, 2021, pp. 3-29. [54] O. Bisikalo, V. Vysotska, Linguistic analysis method of Ukrainian commercial textual content for data mining, volume Vol-2608 of CEUR Workshop Proceedings, 2020, pp. 224-244. [55] B. Rusyn, L. Pohreliuk, A. Rzheuskyi, R. Kubik, Y. Ryshkovets, L. Chyrun, S. Chyrun, A. Vysotskyi, V. B. Fernandes, The Mobile Application Development Based on Online Music Library for Socialising in the World of Bard Songs and Scouts' Bonfires, volume 1080 of Advances in Intelligent Systems and Computing IV, 2020, pp. 734-756.