Comparison of Selected Machine Learning Algorithms in the Analysis of Mental Health Indicators
Abstract
:1. Introduction
Related Publications
- Logistic regression;
- K-nearest-neighbormethod;
- Decision trees;
- Random forest;
- Boosting (increasing the effectiveness of existing models);
- Bagging.
- Classification accuracy;
- False Positive Rate, which indicates how many negative cases were classified as positive;
- Precision, i.e., the fraction of cases predicted to be positive that were actually positive;
- Area Under the Curve (AUC) score;
- Cross-validation AUC score [10].
- Impact of memory problems;
- Possibility to perform only in limited time windows;
- Possibility to perform only under controlled conditions;
- Frequent requirement for the patient to move to a medical setting in order to receive a diagnosis.
- Inability to assess the impact of interaction with the environment in the context of the mental state in real time;
- This undermines progress towards understanding and classifying mental illness and its treatment.
- Fitbit;
- Garmin;
- Healthkit;
- Misfit;
- Moves;
- Myfitnesspal;
- Strava [11].
- Examined daily activity time received from wearable devices was greater than that derived from the mobile phone app;
- Of the 43 participants from whom at least three daily activity observations were obtained, 11 of them had at least 20% missing data between the first and last observation, but this did not show a relationship with DASS-21 scores;
- For the remaining 32 participants, entropy techniques were used, which initially showed no significant relationship between data and DASS-21 scale scores. It was not until splitting into two equal groups in relation to the amount of data that a significant, positive correlation was detected between the DASS-21 anxiety subscale and entropy in those with more data [11].
- Ease of application;
- Possibility to monitor day by day, which conventional methods do not allow.
- Recording a voice from a microphone;
- Analyzing this voice;
- Determining a health indicator based on this.
- 0.795; 0.643; 0.660 for the short-term indicator;
- 1.000; 0.605; 0.646 for the medium-term indicator [12].
- Use of incomplete samples;
- Use of unverified or modified assessment tools;
- Lack of comparable pre-pandemic data to measure change.
- Age;
- Sex;
- Family income;
- Employment status;
- Living with a partner;
- Presence of risk factors [13].
- Higher GHQ-12 scores in women;
- Higher scores in younger age groups;
- Slight differences in ethnicity (apart from the difference between Asians and white British—Asians scored higher);
- Slightly lower results were recorded outside cities;
- Higher scores in low-income families;
- Unemployed and professionally inactive people scored higher than employed and retired people;
- People without a partner and with young children had higher scores, as did the risk groups;
- Significant increase in average scores was noticed comparing the state before and during the pandemic [13].
- Globalization;
- Pressures in the workplace;
- Competition [14].
- Questionnaires;
- Sensors of wearable devices;
- Biological signals [14].
- Educational achievements;
- Socioeconomic achievements;
- Satisfaction with life;
- Quality of interpersonal relations;
- Regression analysis;
- K-nearest neighbors method;
- Decision trees;
- Support vector method;
- Fuzzy logic;
- K-means method [14].
- K-means;
- Hierarchical;
- Based on density;
- And their variants [14].
- Depression;
- Loneliness;
- Fear for your health [16].
- Personality;
- Sex;
- Own results at work;
- Loss of a job by a family member [17].
- Were you a nervous person?
- Have you felt so down that nothing could cheer you up?
- Did you feel calm and composed?
- Have you felt depressed?
- Were you a happy person? [17].
- Patient Health Questionnaire, Adolescent version (PHQ-A);
- Hospital Anxiety and Depression Scale (HADS);
- CRAFFT questionnaire;
- Tobacco Use Questionnaire;
- Rosenberg’s self-esteem scale;
- Kidscreen questionnaire [18].
- Depression;
- Thoughts of suicide;
- Medicines;
- Using alcohol/stimulants;
- Tobacco use;
- Any of the options: about depression, about fears or use of alcohol/stimulants.
- Choice of the ML method affects the regression accuracy, learning time and running time;
- Differences in accuracy are relatively small—up to about 10 percentage points difference between methods.
2. Materials and Methods
2.1. Material
2.2. Methods
- C# language was used to describe the actions performed by the program;
- XAML was used to develop the layout of the user interface in a Universal Windows Platform (UWP) application, along with the naming of elements (which allows them to be used in C# as variables), or the binding of events to specific functions in the code behind the interface (code behind).
- Possibility of streaming learning, i.e., operating on data without having to put it all in memory at once;
- Achieving satisfactory results with a small number of circuits through the entire data set;
- Not wasting computing power on zeros in sparse datasets [24].
- Metrics: mean absolute error, mean squared error, mean squared error, coefficient of determination;
- Learning time, expressed in milliseconds: minimum, average, maximum;
- Prediction time, expressed in milliseconds: minimum, average, maximum.
- Minimum value for learning times, prediction, mean absolute error, mean squared error;
- Maximum value for the coefficient of determination.
- Maximum value for learning times, prediction, mean absolute error, mean squared error, mean squared error,
- Minimum value for the coefficient of determination.
- SDCA: c (regularization strength) and stopping time;
- LBFGS: solver, penalty, max_iter, c, tol, fit_intercept, intercept_scaling, class_weight, random_state, multi_class, verbose, warm_start, and l1_ratio;
- OGD: learning rate and diameter of the decision set.
3. Results
4. Discussion
4.1. Limitations of Studies
- Lack of unequivocal measures of mental health (patients and healthy people)—mental health is a subjective concept and difficult to define unambiguously, which complicates the process of creating ML models;
- Population diversity—individual healthy individuals differ from each other in terms of mental health as well as in different life contexts, which makes general modeling difficult and it will be necessary to adapt models to different population groups;
- Lack of qualitative data—most of the available data is quantitative, which can hinder a fuller understanding of mental health;
- Lack of historical data—it is often important to consider the historical context of the patient’s illness;
- Data privacy—mental health data are very sensitive, so it is necessary to maintain appropriate standards of data privacy and security;
- Cultural differences—Mental health can be understood and experienced differently in different cultures.
- Interpretability of models—understanding why a model made certain decisions can be a problem for mental health diagnosis and treatment;
4.2. Directions for Further Research
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Asatryan, B.; Bleijendaal, H.; Wilde, A.A.M. Toward advanced diagnosis and management of inherited arrhythmia syndromes: Harnessing the capabilities of artificial intelligence and machine learning. Heart Rhythm. 2023, 20, 1399–1407. [Google Scholar] [CrossRef]
- Kannampallil, T.; Dai, R.; Lv, N.; Xiao, L.; Lu, C.; Ajilore, O.A.; Snowden, M.B.; Venditti, E.M.; Williams, L.M.; Kringle, E.A.; et al. Cross-trial prediction of depression remission using problem-solving therapy: A machine learning approach. J. Affect. Disord. 2022, 308, 89–97. [Google Scholar] [CrossRef] [PubMed]
- Hong, N.; Liu, C.; Gao, J.; Han, L.; Chang, F.; Gong, M.; Su, L. State of the Art of Machine Learning-Enabled Clinical Decision Support in Intensive Care Units: Literature Review. JMIR Med. Inform. 2022, 10, e28781. [Google Scholar] [CrossRef]
- Lopez-Jimenez, F.; Attia, Z.; Arruda-Olson, A.M.; Carter, R.; Chareonthaitawee, P.; Jouni, H.; Kapa, S.; Lerman, A.; Luong, C.; Medina-Inojosa, J.R.; et al. Artificial Intelligence in Cardiology: Present and Future. Mayo Clin. Proc. 2020, 95, 1015–1039. [Google Scholar] [CrossRef]
- Reid, J.E.; Eaton, E. Artificial intelligence for pediatric ophthalmology. Curr. Opin. Ophthalmol. 2019, 30, 337–346. [Google Scholar] [CrossRef] [PubMed]
- Mentis, A.A.; Lee, D.; Roussos, P. Applications of artificial intelligence–machine learning for detection ofstress: A critical overview. Mol. Psychiatry 2023, 1–13. [Google Scholar] [CrossRef]
- Galatzer-Levy, I.R.; Onnela, J.P. Machine Learning and the Digital Measurement of Psychological Health. Annu. Rev. Clin. Psychol. 2023, 19, 133–154. [Google Scholar] [CrossRef] [PubMed]
- Sutrisno, S.; Khairina, N.; Syah, R.B.Y.; Eftekhari-Zadeh, E.; Amiri, S. Improved Artificial Neural Network with High Precision for Predicting Burnout among Managers and Employees of Start-Upsduring COVID-19 Pandemic. Electronics 2023, 12, 1109. [Google Scholar] [CrossRef]
- Adapa, K.; Pillai, M.; Foster, M.; Charguia, N.; Mazur, L. Using Explainable Supervised Machine Learning to Predict Burnout in Healthcare Professionals. Stud. Health Technol. Inform. 2022, 294, 58–62. [Google Scholar] [CrossRef]
- Srinivasulu Reddy, U.; Thota, A.; Dharun, A. Machine Learning Techniques for Stress Prediction in Working Employees. In Proceedings of the 2018 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), Madurai, India, 13–15 December 2018; pp. 1–4. [Google Scholar]
- Knight, A.; Bidargaddi, N. Commonly available activity tracker apps and wearables as a mental health outcome indicator: A prospective observational cohort study among young adults with psychological distress. J. Affect. Disord. 2018, 236, 31–36. [Google Scholar] [CrossRef]
- Hagiwara, N. Validity of Mind Monitoring System as a Mental Health Indicator using Voice. Adv. Sci. Technol. Eng. Syst. J. 2017, 2, 338–344. [Google Scholar] [CrossRef]
- Pierce, M. Mental health before and during the COVID-19 pandemic: A longitudinal probability sample survey of the UK population. Lancet Psychiatry 2020, 7, 883–892. [Google Scholar] [CrossRef] [PubMed]
- Srividya, M.; Mohanavalli, S.; Bhalaji, N. Behavioral modeling for mental health using machine learning algorithms. J. Med. Syst. 2018, 42, 88. [Google Scholar] [CrossRef] [PubMed]
- Guo, C.; Tomson, G.; Keller, C.; Söderqvist, F. Prevalence and correlates of positive mental health in Chinese adolescents. BMC Public Health 2018, 18, 263. [Google Scholar] [CrossRef] [PubMed]
- Witteveen, D.; Velthorst, E. Economic hardship and mental health complaints during COVID-19. Proc. Natl. Acad. Sci. USA 2020, 117, 27277–27284. [Google Scholar] [CrossRef]
- Bubonya, M.; Cobb-Clark, D.A.; Wooden, M. Jobloss and the mental health of spouses and adolescent children. IZAJ. LaborEcon. 2017, 6, 6. [Google Scholar]
- Brown, M.R.G. After the Fort McMurray wild fire there are significant increases in mental health symptoms ingrade 7–12 students compared to controls. BMC Pyschiatry 2019, 19, 18. [Google Scholar]
- Pal, S.; Xu, T.; Yang, T.; Rajasekaran, S.; Bi, J. Hybrid-DCA: A double asynchronous approach for stochastic dual coordinate ascent. J. Parallel Distrib. Comput. 2020, 143, 47–66. [Google Scholar] [CrossRef]
- Spiridonoff, A.; Olshevsky, A.; Paschalidis, I.C. Robust Asynchronous Stochastic Gradient-Push: Asymptotically Optimaland Network-Independent Performance for Strongly Convex Functions. J. Mach. Learn. Res. 2020, 21, 58. [Google Scholar]
- Pu, S.; Olshevsky, A.; Paschalidis, I.C. A Sharp Estimate on the Transient Timeoff Distributed Stochastic Gradient Descent. IEEE Trans. Automat. Contr. 2022, 67, 5900–5915. [Google Scholar] [CrossRef]
- Pu, S.; Olshevsky, A.; Paschalidis, I.C. Asymptotic Network Independence in Distributed Stochastic Optimization for Machine Learning. IEEE Signal Process. Mag. 2020, 37, 114–122. [Google Scholar] [CrossRef]
- Mohsen, F.; Al-Saadi, B.; Abdi, N.; Khan, S.; Shah, Z. Artificial Intelligence-Based Methods for Precision Cardiovascular Medicine. J. Pers. Med. 2023, 13, 1268. [Google Scholar] [CrossRef]
- Price, M.J. Hello, C#! Welcome,. NET! In C# 8.0 and.NET Core 3.0—Modern Cross-Platform Development, 4th ed.; Packt Publishing Ltd.: Birmingham, UK, 2019; pp. 1–69. [Google Scholar]
- Perkins, B.; Hammer, J.V.; Reid, J.D. Introducing C#. In Beginning C# 7 Programming with Visual Studio 2017; Wiley: Hoboken, NJ, USA, 2018; pp. 3–13. [Google Scholar]
- Shalev-Shwartz, S.; Tong, Z. Stochastic Dual Coordinate Ascent Methods for Regularized Loss Minimization. arXiv 2013, arXiv:1209.1873. [Google Scholar]
- Lu, X.; Yang, C.; Wu, Q.; Wang, J.; Wei, Y.; Zhang, L.; Li, D.; Zhao, L. Improved Reconstruction Algorithm of Wireless Sensor Network Based on BFGS Quasi-Newton Method. Electronics 2023, 12, 1267. [Google Scholar] [CrossRef]
- Aggrawal, H.O.; Modersitzki, J. Hessian Initialization Strategies for L-BFGS Solving Non-linear Inverse Problems. arXiv 2021, arXiv:2103.10010. [Google Scholar]
- Asl, A.; Overton, M.L. Behavior of limited memory BFGS when applied to nonsmooth functions and their nesterov smoothings. arXiv 2020, arXiv:2006.11336. [Google Scholar]
- Bousbaa, Z.; Sanchez-Medina, J.; Bencharef, O. Financial Time Series Forecasting: A Data Stream Mining-Based System. Electronics 2023, 12, 2039. [Google Scholar] [CrossRef]
- Benczúr, A.A.; Kocsis, L.; Pálovics, R. Online Machine Learning in Big Data Streams. arXiv 2018, arXiv:1802.05872. [Google Scholar]
- Ilboudo, W.E.L.; Kobayashi, T.; Sugimoto, K. Robust stochastic gradient descent with student-t distribution basedfirst-order momentum. IEEE Trans. Neural Netw. Learn. Syst. 2020, 33, 1324–1337. [Google Scholar] [CrossRef]
- Figalová, N.; Charvat, M. The Perceived Stress Scale: Reliability and validity study in the Czech Republic. Ceskoslovenská Psychol. 2021, 65, 46–59. [Google Scholar] [CrossRef]
- Prasetya, A.; Purnama, D.; Prasetyo, F. Validity and Reliability of The Perceived Stress Scale with RASCH Model. PSIKOPEDAGOGIA J. Bimbing. Konseling 2020, 8, 48–51. [Google Scholar] [CrossRef]
- Maslach, C.; Jackson, S.E. The measurement of experienced burnout. J. Occup. Behav. 1981, 2, 99–113. [Google Scholar] [CrossRef]
- Schaufeli, W.B.; Bakker, A.B.; Hoogduin, K.; Kladler, A.; Schaap, C. On the clinical validity of the Maslach Burnout Inventory and the Burnout Measure. Psychol. Health 2001, 16, 565–582. [Google Scholar] [CrossRef] [PubMed]
- Checa, I.; Perales, J.; Espejo, B. Measurement in variance of the Satisfaction with Life Scale by gender, age, marital status and educational level. Qual. Life Res. Int. J. Qual. Life Asp. Treat. Care Rehabil. 2019, 28, 963–968. [Google Scholar] [CrossRef]
- Diener, E.; Emmons, R.A.; Larsen, R.J.; Griffin, S. The Satisfaction with Life Scale. J. Personal. Assess. 1985, 49, 71–75. [Google Scholar] [CrossRef]
- Prokopowicz, P.; Mikołajewski, D.; Mikołajewska, E. Intelligent System for Detecting Deterioration of Life Satisfaction as Tool for Remote Mental-Health Monitoring. Sensors 2022, 22, 9214. [Google Scholar] [CrossRef]
- Rojek, I. Neural networks as prediction models for water intake in water supply system. In Artificial Intelligence and Soft Computing—ICAISC 2008. Lecture Notes in Computer Science, 5097; Rutkowski, L., Tadeusiewicz, R., Zadeh, L.A., Zurada, J.M., Eds.; Springer: Berlin/Heidelberg, Gemany, 2008; pp. 1109–1119. Available online: https://link.springer.com/chapter/10.1007/978-3-540-69731-2_104 (accessed on 31 August 2023).
- Spoor, J.M.; Weber, J. Evaluation of process planning in manufacturing by a neural network based on an energy definition of Hopfield nets. J. Intell. Manuf. 2023, 1–19. [Google Scholar] [CrossRef]
- Teixeira, I.; Morais, R.; Sousa, J.J.; Cunha, A. Deep Learning Models for the Classification of Cropsin Aerial Imagery: A Review. Agriculture 2023, 13, 965. [Google Scholar] [CrossRef]
- Rojek, I.; Mikołajewski, D.; Macko, M.; Szczepański, Z.; Dostatni, E. Optimization of Extrusion-Based 3D Printing Process Using Neural Networks for Sustainable Development. Materials 2021, 14, 2737. [Google Scholar] [CrossRef]
- Rojek, I.; Mikołajewski, D.; Kotlarz, P.; Macko, M.; Kopowski, J. Intelligent system supporting technological process planning for machining and 3D printing. Bull. Pol. Acad. Sci. Tech. Sci. 2021, 69, e136722. [Google Scholar]
- Mohammadi, E.K.; Talaie, H.R.; Azizi, M. A healthcare service quality assessment model usinga fuzzy best–worst method with application to hospitals within-patient services. Healthc. Anal. 2023, 4, 100241. [Google Scholar] [CrossRef]
- Gajos, A.; Wójcik, G.M. Independent component analysis of EEG data for EGI system. Bio-Algorithms Med-Syst. 2016, 12, 67–72. [Google Scholar] [CrossRef]
- Kawala-Janik, A.; Podpora, M.; Pelc, M.; Piatek, P.; Baranowski, J. Implementation of an inexpensive EEG headset for the pattern recognition purpose. In Proceedings of the 2013 IEEE 7th International Conference on Intelligent Data Acquisition and Advanced Computing Systems (IDAACS), Berlin, Germany, 12–14 September 2013; Volume 1, pp. 399–403. [Google Scholar]
Parameter | Mean | SD | Min | Q1 | Median | Q3 | Max |
---|---|---|---|---|---|---|---|
PSS item 1 | 2.96 | 0.79 | 1 | 2 | 3 | 4 | 4 |
PSS item 2 | 3.14 | 0.74 | 2 | 3 | 3 | 4 | 4 |
PSS item 3 | 2.87 | 0.92 | 1 | 2 | 3 | 4 | 4 |
PSS item 4 | 2.66 | 1.05 | 0 | 2 | 3 | 3 | 4 |
PSS item 5 | 3.06 | 0.65 | 1 | 3 | 3 | 3 | 4 |
PSS item 6 | 2.90 | 0.85 | 1 | 2 | 3 | 3 | 4 |
PSS item 7 | 3.08 | 0.97 | 1 | 3 | 3 | 4 | 4 |
PSS item 8 | 2.67 | 0.90 | 0 | 2 | 3 | 3 | 4 |
PSS item 9 | 2.94 | 0.71 | 1 | 3 | 3 | 3 | 4 |
PSS item 10 | 2.49 | 0.93 | 1 | 2 | 2 | 3 | 4 |
MBI item 1 | 3.27 | 1.96 | 0 | 2 | 3 | 5 | 6 |
MBI item 2 | 2.73 | 1.73 | 0 | 2 | 3 | 4 | 6 |
MBI item 3 | 2.49 | 1.70 | 0 | 1 | 3 | 3 | 5 |
MBI item 4 | 2.24 | 2.32 | 0 | 0 | 1 | 5 | 6 |
MBI item 5 | 1.50 | 1.68 | 0 | 0 | 1 | 3 | 6 |
MBI item 6 | 1.53 | 1.48 | 0 | 0 | 1 | 3 | 6 |
MBI item 7 | 3.37 | 1.78 | 0 | 2 | 3 | 5 | 6 |
MBI item 8 | 1.69 | 1.68 | 0 | 0 | 1 | 3 | 6 |
MBI item 9 | 2.86 | 2.57 | 0 | 0 | 3 | 6 | 6 |
MBI item 10 | 1.56 | 1.35 | 0 | 1 | 1 | 3 | 6 |
MBI item 11 | 2.09 | 1.55 | 0 | 0 | 3 | 3 | 6 |
MBI item 12 | 2.55 | 1.66 | 0 | 1 | 3 | 3 | 6 |
MBI item 13 | 2.09 | 1.52 | 0 | 1 | 2 | 3 | 6 |
MBI item 14 | 2.17 | 1.86 | 0 | 1 | 1 | 3 | 6 |
MBI item 15 | 2.36 | 2.03 | 0 | 0 | 2 | 4 | 6 |
MBI item 16 | 1.68 | 1.77 | 0 | 0 | 1 | 2.5 | 6 |
MBI item 17 | 2.76 | 2.04 | 0 | 1 | 3 | 3 | 6 |
MBI item 18 | 1.64 | 1.45 | 0 | 0 | 1 | 3 | 5 |
MBI item 19 | 2.56 | 1.95 | 0 | 1 | 3 | 3 | 6 |
MBI item 20 | 1.87 | 2.23 | 0 | 0 | 1 | 3 | 6 |
MBI item 21 | 1.21 | 1.48 | 0 | 0 | 0 | 3 | 4 |
MBI item 22 | 2.43 | 1.45 | 0 | 2 | 3 | 4 | 6 |
SWLS item 1 | 4.08 | 0.93 | 2 | 4 | 4 | 5 | 5 |
SWLS item 2 | 3.24 | 1.53 | 1 | 2 | 3 | 4 | 6 |
SWLS item 3 | 3.30 | 1.66 | 1 | 1 | 4 | 5 | 6 |
SWLS item 4 | 3.20 | 1.66 | 1 | 2 | 2 | 4 | 6 |
SWLS item 5 | 2.51 | 1.59 | 1 | 1 | 2 | 4 | 5 |
Parameter | Micro Accuracy (%) | Macro Accuracy (%) | Best Trainer |
---|---|---|---|
Gender | 75.16 | 69.32 | FastTreeOva |
Age | 71.24 | 62.82 | FastTreeOva |
Seniority | 78.73 | 72.23 | FastForestOva |
Total pts. | 17.29 | 14.79 | LightGbmMulti |
Parameter | Accuracy (%) | Best Trainer |
---|---|---|
Gender | Not possible | |
Age | 93.32 | LbfgsPoissonRegressionRegression |
Seniority | 97.57 | FastTreeRegression |
Total pts. | 97.42 | LbfgsPoissonRegressionRegression |
Area | Description and Detailed Tasks |
---|---|
Data collection and analysis | The use of many different data sources, including multi-modal ones, such as behavioral data (e.g., online activity, phone calls), biometric data (e.g., heart rate, sleep monitoring), survey data, photos and videos, as well as test results collected automatically, etc. |
Collaboration with field experts | Collaboration with physicians and mental health professionals can help understand the mechanisms and create and evaluate the effectiveness of models. |
Ethics and privacy | The manner in which data are collected, stored, used and destroyed should comply with relevant regulations and ethical standards. |
Data preparation | Data preparation may include data normalization, removal of erroneous, uncertain, incomplete and outlier data, coding of categorical variables, etc. |
Selection of ML algorithms and hyperparameters | Selection and adaptation of algorithms and hyperparameters of models to a specific problem from among possible solutions, such as decision trees, neural networks, support vector machines (SVM) or clustering algorithms. |
Evaluation/cross-validation of models | Define model performance metrics (accuracy, sensitivity, specificity, F1-score, Receiver Operating Characteristic (ROC) curves, etc.) and analyze model performance using them. |
Interpretability of models | Understanding how the model makes its predictions (why the model made certain decisions). |
Checking the learning time | Model training time can be a critical factor in clinical practice—it needs to be investigated how long it takes to train different models and whether this can be optimized. |
The model should be adapted to real-time operation (including learning on new patients) in order to be used in clinical practice. | |
Validation on a large sample of patients | The effectiveness of the models should be tested on a large sample of patients to ensure that the model generalizes well to different cases. |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Bieliński, A.; Rojek, I.; Mikołajewski, D. Comparison of Selected Machine Learning Algorithms in the Analysis of Mental Health Indicators. Electronics 2023, 12, 4407. https://doi.org/10.3390/electronics12214407
Bieliński A, Rojek I, Mikołajewski D. Comparison of Selected Machine Learning Algorithms in the Analysis of Mental Health Indicators. Electronics. 2023; 12(21):4407. https://doi.org/10.3390/electronics12214407
Chicago/Turabian StyleBieliński, Adrian, Izabela Rojek, and Dariusz Mikołajewski. 2023. "Comparison of Selected Machine Learning Algorithms in the Analysis of Mental Health Indicators" Electronics 12, no. 21: 4407. https://doi.org/10.3390/electronics12214407