A Novel Integration of IF-DEMATEL and TOPSIS for the Classifier Selection Problem in Assistive Technology Adoption for People with Dementia
Abstract
:1. Introduction
- ▪
- The IF-DEMATEL technique allows for detecting and measuring potential cause-effect interrelations among decision criteria involved in technology adoption. Its inclusion lays the groundwork for the identification of the main drivers underpinning the effective design, development, and implementation of classification algorithms in the real world. In addition, it considers the vagueness and uncertainty inherent in human judgments, an aspect of paramount importance to take into account from the participation of different stakeholders whose expectations are desired to be fully incorporated into the classifier selection model [10,11]. Another contribution from this method is the possibility to estimate the importance of conflicting criteria and sub-criteria with respect to the goal; in this case, the identification of the most suitable classifier supporting technology adoption in PwD.
- ▪
- It is straightforward to apply the TOPSIS method in the wild (physical, environmental, and organisational elements) and it is employed for ranking classifier alternatives based on a data-driven approach generating a closeness coefficient (in this case, the suitability index). One of the main weaknesses to be overcome in this technique is the allocation of weights that are originally assigned randomly [7,12]. The inclusion of IF-DEMATEL tackles this disadvantage by providing a solid mathematical foundation considering interdependences and feedback among the decision elements.
2. Related Work
2.1. A Background on Assistive Technology Solutions (ATS) for PwD
2.2. Technology Adoption in PwD from an MCDM Perspective
3. Proposed Methodology
3.1. Intuitionistic Fuzzy Decision Making Trial and Evaluation Laboratory (IF-DEMATEL)
3.2. Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS)
- Construct the performance matrix by considering classifiers and decision elements. Each denotes the value of the decision element in each classifier.
- Obtain the normalized performance matrix using Equation (16) where is the modified TOPSIS norm suggested by García-Cascales and Lamata [62]. Estimate via utilizing Equation (17). symbolizes the element of the normalized performance matrix corresponding to the decision criterion of each classifier .
- Compute the weighted normalized performance matrix via implementing Equation (18). The sub-criteria priorities are given by the IF-DEMATEL technique.
- Establish absolute positive and negative ideal solutions according to García-Cascales and Lamata [62].
- Obtain the Euclidean distance from the positive (PIS) and negative (NIS) ideal solutions for each classifier via utilizing Equations (19) and (20) correspondingly.
- Estimate the relative closeness coefficient to the ideal solution (that is, the “suitability index”) of each classifier via Equation (21).
- Rank the classifiers in a decreasing order based on values.
4. A Case Study of a Mobile-Based Reminding Solution
4.1. The Decision-Making Group
4.2. The Classifier Selection Network
4.3. Calculation of Fuzzy Relative Priorities and Interdependence Evaluation: The IF-DEMATEL Approach
4.4. Calculation of Suitability Index per Classifier and Detection of Improvement Opportunities: The Modified TOPSIS Approach
4.5. Validation Study: Contrasting TOPSIS Results with VIKOR and SAW
5. Concluding Remarks and Future Directions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
Appendix A
SF9 | SF10 | SF11 | ||||
SF9 | 0 | 0 | 0.90 | 0.10 | 0.50 | 0.45 |
SF10 | 0.50 | 0.45 | 0 | 0 | 0.90 | 0.10 |
SF11 | 0.50 | 0.45 | 0.10 | 0.90 | 0 | 0 |
SF9 | SF10 | SF11 | |
SF9 | 0 | 0.90 | 0.53 |
SF10 | 0.53 | 0 | 0.90 |
SF11 | 0.53 | 0.10 | 0 |
SF9 | SF10 | SF11 | |
SF9 | 0 | 3.60 | 2.10 |
SF10 | 2.10 | 0 | 3.60 |
SF11 | 2.10 | 0.40 | 0 |
SF9 | SF10 | SF11 | |
SF9 | 0 | 2.18 | 1.91 |
SF10 | 2.65 | 0 | 2.23 |
SF11 | 2.66 | 2.45 | 0 |
SF9 | SF10 | SF11 | |
SF9 | 0 | 0.409 | 0.360 |
SF10 | 0.499 | 0 | 0.419 |
SF11 | 0.501 | 0.461 | 0 |
SF9 | SF10 | SF11 | D | |
SF9 | 2.184 | 2.270 | 2.097 | 6.551 |
SF10 | 2.796 | 2.234 | 2.361 | 7.391 |
SF11 | 2.885 | 2.629 | 2.140 | 7.654 |
R | 7.865 | 7.133 | 6.598 |
SF1 | SF2 | SF3 | SF4 | SF5 | SF6 | SF7 | SF8 | F3 | |
A1 | 0.85 | 1 | 0.664 | 0.584 | 0.717 | 0.61 | 0 | 0 | 1 |
A2 | 0.85 | 1 | 0.504 | 0.637 | 0.637 | 0.425 | 0 | 1 | 1 |
A3 | 0.825 | 1 | 0.478 | 0.212 | 0.239 | 0.557 | 0 | 0 | 0 |
A4 | 0.349 | 1 | 0.239 | 0.584 | 0.504 | 0.159 | 1 | 1 | 0 |
A5 | 0.875 | 2 | 0.185 | 0.239 | 0.212 | 0.159 | 0 | 0 | 0 |
A6 | 0.825 | 5 | 0.371 | 0.132 | 0.132 | 0.504 | 1 | 1 | 0 |
A7 | 0.825 | 1 | 0.212 | 0.265 | 0.239 | 0.265 | 1 | 1 | 1 |
A+ | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
A- | 0 | 5 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
W | 0.035 | 0.029 | 0.032 | 0.033 | 0.033 | 0.033 | 0.102 | 0.102 | 0.185 |
SF9 | SF10 | SF11 | SF12 | SF13 | SF14 | SF15 | SF16 | ||
A1 | 1 | 0 | 1 | 1 | 1 | 2 | 1 | 0 | |
A2 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | |
A3 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | |
A4 | 1 | 1 | 1 | 1 | 0 | 2 | 1 | 1 | |
A5 | 1 | 1 | 1 | 0 | 0 | 2 | 1 | 1 | |
A6 | 1 | 1 | 1 | 1 | 0 | 2 | 1 | 1 | |
A7 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | |
A+ | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | |
A- | 0 | 0 | 0 | 0 | 1 | 2 | 0 | 0 | |
W | 0.066 | 0.067 | 0.066 | 0.044 | 0.04 | 0.044 | 0.045 | 0.043 |
SF1 | SF2 | SF3 | SF4 | SF5 | SF6 | SF7 | SF8 | F3 | |
A1 | 0.00050 | 0.00021 | 0.00062 | 0.00075 | 0.00060 | 0.00072 | 0.01040 | 0.01040 | 0.00856 |
A2 | 0.00009 | 0.00000 | 0.00057 | 0.00038 | 0.00038 | 0.00073 | 0.01040 | 0.00000 | 0.00000 |
A3 | 0.00012 | 0.00000 | 0.00061 | 0.00099 | 0.00097 | 0.00052 | 0.01040 | 0.01040 | 0.03423 |
A4 | 0.00108 | 0.00021 | 0.00097 | 0.00075 | 0.00083 | 0.00106 | 0.00260 | 0.00260 | 0.03423 |
A5 | 0.00047 | 0.00084 | 0.00099 | 0.00103 | 0.00104 | 0.00106 | 0.01040 | 0.01040 | 0.03423 |
A6 | 0.00091 | 0.01346 | 0.00097 | 0.00108 | 0.00108 | 0.00098 | 0.00666 | 0.00666 | 0.03423 |
A7 | 0.00012 | 0.00000 | 0.00093 | 0.00094 | 0.00097 | 0.00094 | 0.00000 | 0.00000 | 0.00000 |
SF9 | SF10 | SF11 | SF12 | SF13 | SF14 | SF15 | SF16 | Si+ | |
A1 | 0.00109 | 0.00449 | 0.00109 | 0.00048 | 0.00040 | 0.00194 | 0.00051 | 0.00185 | 0.21121 |
A2 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00160 | 0.00000 | 0.00000 | 0.00185 | 0.12656 |
A3 | 0.00436 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.25021 |
A4 | 0.00109 | 0.00112 | 0.00109 | 0.00048 | 0.00000 | 0.00194 | 0.00051 | 0.00046 | 0.22586 |
A5 | 0.00109 | 0.00112 | 0.00109 | 0.00194 | 0.00000 | 0.00194 | 0.00051 | 0.00046 | 0.26192 |
A6 | 0.00279 | 0.00287 | 0.00279 | 0.00124 | 0.00000 | 0.00008 | 0.00130 | 0.00118 | 0.27977 |
A7 | 0.00000 | 0.00000 | 0.00000 | 0.00194 | 0.00160 | 0.00000 | 0.00000 | 0.00000 | 0.08629 |
SF1 | SF2 | SF3 | SF4 | SF5 | SF6 | SF7 | SF8 | F3 | |
A1 | 0.00016 | 0.01703 | 0.00005 | 0.00003 | 0.00007 | 0.00004 | 0.00000 | 0.00000 | 0.00856 |
A2 | 0.00064 | 0.01346 | 0.00007 | 0.00018 | 0.00018 | 0.00004 | 0.00000 | 0.01040 | 0.03423 |
A3 | 0.00057 | 0.01346 | 0.00005 | 0.00000 | 0.00000 | 0.00010 | 0.00000 | 0.00000 | 0.00000 |
A4 | 0.00000 | 0.01703 | 0.00000 | 0.00003 | 0.00002 | 0.00000 | 0.00260 | 0.00260 | 0.00000 |
A5 | 0.00018 | 0.00757 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00000 |
A6 | 0.00002 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00000 | 0.00042 | 0.00042 | 0.00000 |
A7 | 0.00057 | 0.01346 | 0.00000 | 0.00001 | 0.00000 | 0.00001 | 0.01040 | 0.01040 | 0.03423 |
SF9 | SF10 | SF11 | SF12 | SF13 | SF14 | SF15 | SF16 | Si- | |
A1 | 0.00109 | 0.00000 | 0.00109 | 0.00048 | 0.00014 | 0.00279 | 0.00051 | 0.00000 | 0.17899 |
A2 | 0.00436 | 0.00449 | 0.00436 | 0.00194 | 0.00102 | 0.00008 | 0.00203 | 0.00000 | 0.27829 |
A3 | 0.00000 | 0.00449 | 0.00436 | 0.00194 | 0.00006 | 0.00008 | 0.00203 | 0.00185 | 0.17025 |
A4 | 0.00109 | 0.00112 | 0.00109 | 0.00048 | 0.00006 | 0.00279 | 0.00051 | 0.00046 | 0.17289 |
A5 | 0.00109 | 0.00112 | 0.00109 | 0.00000 | 0.00006 | 0.00279 | 0.00051 | 0.00046 | 0.12195 |
A6 | 0.00017 | 0.00018 | 0.00017 | 0.00008 | 0.00006 | 0.00000 | 0.00008 | 0.00007 | 0.04103 |
A7 | 0.00436 | 0.00449 | 0.00436 | 0.00000 | 0.00102 | 0.00008 | 0.00203 | 0.00185 | 0.29538 |
References
- Wittenberg, R.; Hu, B.; Barraza-Araiza, L.; Rehill, A. Projections of Older People with Dementia and Costs of Dementia Care in the United Kingdom, 2019–2040; London School of Economics: London, UK, 2019. [Google Scholar]
- Dawson, W.D.; Bangerter, L.R.; Splaine, M. The politics of caregiving: Taking stock of state-level policies to support family caregivers. Public Policy Aging Rep. 2020, 30, 62–66. [Google Scholar] [CrossRef]
- Ortiz-Barrios, M.; Nugent, C.; Cleland, I.; Donnelly, M.; Verikas, A. Selecting the most suitable classification algorithm for supporting assistive technology adoption for people with dementia: A multicriteria framework. J. Multi-Criteria Decis. Anal. 2020, 27, 20–38. [Google Scholar] [CrossRef]
- Siksnelyte-Butkiene, I.; Zavadskas, E.K.; Streimikiene, D. Multi-criteria decision-making (MCDM) for the assessment of renewable energy technologies in a household: A review. Energies 2020, 13, 1164. [Google Scholar] [CrossRef] [Green Version]
- Ejegwa, P.A.; Akowe, S.O.; Otene, P.M.; Ikyule, J.M. An overview on intuitionistic fuzzy sets. Int. J. Sci. Technol. Res. 2014, 3, 142–145. [Google Scholar]
- Si, S.-L.; You, X.-Y.; Liu, H.-C.; Zhang, P. DEMATEL technique: A systematic review of the state-of-the-art literature on methodologies and applications. Math. Probl. Eng. 2018, 2018, 3696457. [Google Scholar] [CrossRef] [Green Version]
- Behzadian, M.; Khanmohammadi Otaghsara, S.; Yazdani, M.; Ignatius, J. A state-of the-art survey of TOPSIS applications. Expert Syst. Appl. 2012, 39, 13051–13069. [Google Scholar] [CrossRef]
- Ocampo, L.; Deiparine, C.B.; Go, A.L. Mapping strategy to best practices for sustainable food manufacturing using fuzzy dematel-anp-topsis. EMJ Eng. Manag. J. 2020, 32, 130–150. [Google Scholar] [CrossRef]
- Sumrit, D. A hybrid multi-criteria decision making model for technological innovation capabilities measurement in automotive parts industry. Int. J. Manag. Decis. Mak. 2020, 19, 1–43. [Google Scholar] [CrossRef]
- Hinduja, A.; Pandey, M. An integrated intuitionistic fuzzy MCDM approach to select cloud-based ERP system for SMEs. Int. J. Inf. Technol. Decis. Mak. 2019, 18, 1875–1908. [Google Scholar] [CrossRef]
- Mishra, A.R.; Mardani, A.; Rani, P.; Zavadskas, E.K. A novel EDAS approach on intuitionistic fuzzy set for assessment of health-care waste disposal technology using new parametric divergence measures. J. Clean. Prod. 2020, 272, 122807. [Google Scholar] [CrossRef]
- Çelikbilek, Y.; Tüysüz, F. An in-depth review of theory of the TOPSIS method: An experimental analysis. J. Manag. Anal. 2020, 7, 281–300. [Google Scholar] [CrossRef]
- Hurst, A.; Tobias, J. Empowering Individuals with do-it-yourself Assistive Technology. In Proceedings of the 13th International ACM SIGACCESS Conference on Computers and Accessibility, Dundee, Scotland, UK, 24–26 October 2011; pp. 11–18. [Google Scholar]
- Lee, C.; Coughlin, J. PERSPECTIVE: Older adults’ adoption of technology: An integrated approach to identifying determinants and barriers. J. Prod. Innov. Manag. 2015, 32, 747–759. [Google Scholar] [CrossRef]
- Fotteler, M.; Risch, B.; Gaugisch, P.; Furmanek, J.L.; Swoboda, W.; Mayer, S.; Kohn, B.; Dallmeier, D.; Denkinger, M. Obstacles to Using Assistive Technology for Older Adults–Results from a Focus Group Analysis. Stud. Health Technol. Inform. 2021, 281, 994–998. [Google Scholar] [PubMed]
- Pal, J.; Viswanathan, A.; Chandra, P.; Nazareth, A.; Kameswaran, V.; Subramonyam, H.; Johri, A.; Ackerman, M.; O’Modhrain, S. Agency in Assistive Technology Adoption: Visual Impairment and Smartphone Use in Bangalore. In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems, Denver, CO, USA, 6–11 May 2017; pp. 5929–5940. [Google Scholar]
- Kintsch, A.; DePaula, R. A Framework for the Adoption of Assistive Technology. SWAAAC 2002: Supporting Learning through Assistive Technology 2002. pp. 1–10. Available online: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.124.3726&rep=rep1&type=pdf (accessed on 1 November 2021).
- Goodman, G.; Tiene, D.; Luft, P. Adoption of assistive technology for computer access among college students with disabilities. Disabil. Rehabil. 2002, 24, 80–92. [Google Scholar] [CrossRef] [PubMed]
- Cleland, I.; Nugent, C.; McClean, S.; Hartin, P.; Sanders, C.; Donnelly, M.; Zhang, S.; Scotney, B.; Smith, K.; Norton, M.C.; et al. Predicting Technology Adoption in People with Dementia; Initial Results from the TAUT Project. In International Workshop on Ambient Assisted Living; Springer: Cham, Switzerland, 2014; pp. 266–274. [Google Scholar]
- Chaurasia, P.; McClean, S.; Nugent, C.; Cleland, I.; Zhang, S.; Donnelly, M.; Scotney, B.; Sanders, C.; Smith, K.; Norton, M.; et al. Technology Adoption and Prediction Tools for Everyday Technologies Aimed at People with Dementia. In Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Orlando, FL, USA, 16–20 August 2016; pp. 4407–4410. [Google Scholar]
- Chaurasia, P.; McClean, S.; Nugent, C.; Cleland, I.; Zhang, S.; Donnelly, M.; Scotney, B.; Sanders, C.; Smith, K.; Norton, M.; et al. Modelling mobile-based technology adoption among people with dementia. Pers. Ubiquitous Comput. 2021, 1–20. [Google Scholar] [CrossRef]
- Ortiz-Barrios, M.; Miranda-De la Hoz, C.; López-Meza, P.; Petrillo, A.; De Felice, F. A case of food supply chain management with AHP, DEMATEL, and TOPSIS. J. Multi-Criteria Decis. Anal. 2020, 27, 104–128. [Google Scholar] [CrossRef]
- Cruz-Sandoval, D.; Favela, J.; Lopez-Nava, I.; Morales, A. Adoption of Wearable Devices by Persons with Dementia: Lessons from a Non-pharmacological Intervention Enabled by a Social Robot. In IoT in Healthcare and Ambient Assisted Living; Springer: Singapore, 2021; pp. 145–163. [Google Scholar]
- Zhang, S.; McClean, S.I.; Nugent, C.D.; Donnelly, M.P.; Galway, L.; Scotney, B.W.; Cleland, I. A predictive model for assistive technology adoption for people with dementia. IEEE J. Biomed. Health Inform. 2014, 18, 375–383. [Google Scholar] [CrossRef] [PubMed]
- Øksnebjerg, L.; Janbek, J.; Woods, B.; Waldemar, G. Assistive technology designed to support self-management of people with dementia: User involvement, dissemination, and adoption. A scoping review. Int. Psychogeriatr. 2020, 32, 937–953. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Øksnebjerg, L.; Woods, B.; Vilsen, C.; Ruth, K.; Gustafsson, M.; Ringkobing, S.; Waldemar, G. Self-management and cognitive rehabilitation in early stage dementia–merging methods to promote coping and adoption of assistive technology. A pilot study. Aging Ment. Health 2020, 24, 1894–1903. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Köksalan, M.; Wallenius, J.; Zionts, S. An early history of multiple criteria decision making. Data Envel. Anal. 2013, 20, 3–17. [Google Scholar] [CrossRef]
- Zavadskas, E.K.; Turskis, Z.; Kildiene, S. State of art surveys of overviews on MCDM/MADM methods. Technol. Econ. Dev. Econ. 2014, 20, 165–179. [Google Scholar] [CrossRef] [Green Version]
- Ortiz-Barrios, M.; Nugent, C.; Garcia-Constantino, M.; Jimenez-Delgado, G. Identifying the Most Appropriate Classifier for Underpinning Assistive Technology Adoption for People With Dementia: An Integration of Fuzzy Ahp And Vikor Methods. In International Conference on Human-Computer Interaction; Springer: Cham, Switzerland, 2020; pp. 406–419. [Google Scholar]
- Zavadskas, E.K.; Govindan, K.; Antucheviciene, J.; Turskis, Z. Hybrid multiple criteria decision-making methods: A review of applications for sustainability issues. Econ. Res.-Ekon. Istraz. 2016, 29, 857–887. [Google Scholar] [CrossRef] [Green Version]
- Ortiz-Barrios, M.; Gul, M.; López-Meza, P.; Yucesan, M.; Navarro-Jiménez, E. Evaluation of hospital disaster preparedness by a multi-criteria decision making approach: The case of turkish hospitals. Int. J. Disaster Risk Reduct. 2020, 49, 101748. [Google Scholar] [CrossRef]
- Faizi, S.; Sałabun, W.; Nawaz, S.; Rehman, A.U.; Wątróbski, J. Best-worst method and hamacher aggregation operations for intuitionistic 2-tuple linguistic sets. Expert Syst. Appl. 2021, 181, 115088. [Google Scholar] [CrossRef]
- Faizi, S.; Sałabun, W.; Rashid, T.; Zafar, S.; Watróbski, J. Intuitionistic fuzzy sets in multi-criteria group decision making problems using the characteristic objects method. Symmetry 2020, 12, 1382. [Google Scholar] [CrossRef]
- Liu, Y.; Eckert, C.M.; Earl, C. A review of fuzzy AHP methods for decision-making with subjective judgements. Expert Syst. Appl. 2020, 161, 113738. [Google Scholar] [CrossRef]
- Büyüközkan, G.; Havle, C.A.; Feyzioğlu, O. A new digital service quality model and its strategic analysis in aviation industry using interval-valued intuitionistic fuzzy AHP. J. Air Transp. Manag. 2020, 86, 101817. [Google Scholar] [CrossRef]
- Hanine, M.; Boutkhoum, O.; Barakaz, F.E.; Lachgar, M.; Assad, N.; Rustam, F.; Ashraf, I. An intuitionistic fuzzy approach for smart city development evaluation for developing countries: Moroccan context. Mathematics 2021, 9, 2668. [Google Scholar] [CrossRef]
- Ocampo, L.; Yamagishi, K. Modeling the lockdown relaxation protocols of the philippine government in response to the COVID-19 pandemic: An intuitionistic fuzzy DEMATEL analysis. Socio-Econ. Plan. Sci. 2020, 72, 100911. [Google Scholar] [CrossRef]
- Otay, İ.; Oztaysi, B.; Cevik Onar, S.; Kahraman, C. Multi-expert performance evaluation of healthcare institutions using an integrated intuitionistic fuzzy AHP&DEA methodology. Knowl. Based Syst. 2017, 133, 90–106. [Google Scholar] [CrossRef]
- Mishra, A.R.; Rani, P.; Mardani, A.; Pardasani, K.R.; Govindan, K.; Alrasheedi, M. Healthcare evaluation in hazardous waste recycling using novel interval-valued intuitionistic fuzzy information based on complex proportional assessment method. Comput. Ind. Eng. 2020, 139, 106140. [Google Scholar] [CrossRef]
- Kalender, Z.T.; Tozan, H.; Vayvay, O. Prioritization of medical errors in patient safety management: Framework using interval-valued intuitionistic fuzzy sets. Healthcare 2020, 8, 265. [Google Scholar] [CrossRef]
- Paradowski, B.; Shekhovtsov, A.; Sałabun, W.; Baczkiewicz, A.; Kizielewicz, B. Similarity analysis of methods for objective determination of weights in multi-criteria decision support systems. Symmetry 2021, 13, 1874. [Google Scholar] [CrossRef]
- Zareravasan, A.; Alizadeh, R. Challenges in creating business value from health information systems (HIS): A hybrid fuzzy approach. J. Inf. Technol. Manag. 2021, 13, 51–74. [Google Scholar] [CrossRef]
- Liu, H.C.; You, J.X.; Zhen, L.; Fan, X.J. A novel hybrid multiple criteria decision making model for material selection with target-based criteria. Mater. Des. 2014, 60, 380–390. [Google Scholar] [CrossRef]
- Ortiz-Barrios, M.A.; Herrera-Fontalvo, Z.; Rúa-Muñoz, J.; Ojeda-Gutiérrez, S.; De Felice, F.; Petrillo, A. An integrated approach to evaluate the risk of adverse events in hospital sector: From theory to practice. Manag. Decis. 2018, 56, 2187–2224. [Google Scholar] [CrossRef] [Green Version]
- Sałabun, W.; Watróbski, J.; Shekhovtsov, A. Are MCDA methods benchmarkable? A comparative study of TOPSIS, VIKOR, COPRAS, and PROMETHEE II methods. Symmetry 2020, 12, 1549. [Google Scholar] [CrossRef]
- Bertolini, M.; Esposito, G.; Romagnoli, G. A TOPSIS-based approach for the best match between manufacturing technologies and product specifications. Expert Syst. Appl. 2020, 159, 113610. [Google Scholar] [CrossRef]
- Durak, İ.; Arslan, H.M.; Özdemir, Y. Application of AHP–TOPSIS methods in technopark selection of technology companies: Turkish case. Technol. Anal. Strateg. Manag. 2021. [Google Scholar] [CrossRef]
- Velasquez, M.; Hester, P.T. An analysis of multi-criteria decision making methods. Int. J. Oper. Res. 2013, 10, 56–66. [Google Scholar]
- Navarro, I.J.; Yepes, V.; Martí, J. A review of multicriteria assessment techniques applied to sustainable infrastructure design. Adv. Civ. Eng. 2019, 2019, 6134803. [Google Scholar] [CrossRef]
- Omar, M.; Hasan, B.; Ahmad, M.; Yasin, A.; Baharom, F.; Mohd, H.; Darus, N.M. Applying fuzzy technique in software team formation based on belbin team role. J. Telecommun. Electron. Comput. Eng. 2016, 8, 109–113. [Google Scholar]
- Dinçer, H.; Yüksel, S. Financial Sector-Based Analysis of the G20 Economies Using the Integrated Decision-Making Approach with DEMATEL and TOPSIS. In Emerging Trends in Banking and Finance; Springer: Cham, Switzerland, 2018; pp. 210–223. [Google Scholar]
- Yalcin, A.; Kilic, H.; Guler, E. Research and Development Project Selection via IF-DEMATEL and IF-TOPSIS. In International Conference on Intelligent and Fuzzy Systems; Springer: Cham, Switzerland, 2019; pp. 625–633. [Google Scholar]
- Zhang, X.; Su, J. A combined fuzzy DEMATEL and TOPSIS approach for estimating participants in knowledge-intensive crowdsourcing. Comput. Ind. Eng. 2019, 137, 106085. [Google Scholar] [CrossRef]
- Erkal, G.; Kilic, H.; Kalender, Z.; Yalcin, A.; Tuzkaya, G. An Integrated IVIF-DEMATEL and IVIF-TOPSIS Methodology for Hotel Information System Selection. In International Conference on Intelligent and Fuzzy Systems; Springer: Cham, Switzerland, 2020; pp. 381–389. [Google Scholar]
- Li, X.; Han, Z.; Zhang, R.; Zhang, Y.; Zhang, L. Risk assessment of hydrogen generation unit considering dependencies using integrated DEMATEL and TOPSIS approach. Int. J. Hydrogen Energy 2020, 45, 29630–29642. [Google Scholar] [CrossRef]
- Atanassov, K.T. Intuitionistic fuzzy sets. In Physica; Springer: Heidelberg, Germany, 1999; pp. 1–137. [Google Scholar]
- Gan, J.; Luo, L. Using DEMATEL and intuitionistic fuzzy sets to identify critical factors influencing the recycling rate of end-of-life vehicles in china. Sustainability 2017, 9, 1873. [Google Scholar] [CrossRef]
- Kilic, H.S.; Demirci, A.E.; Delen, D. An integrated decision analysis methodology based on IF-DEMATEL and IF-ELECTRE for personnel selection. Decis. Support Syst. 2020, 137, 113360. [Google Scholar] [CrossRef]
- Anzilli, L.; Facchinetti, G. A New Proposal of Defuzzification of Intuitionistic fuzzy Quantities. In Novel Developments in Uncertainty Representation and Processing; Springer: Cham, Switzerland, 2016; pp. 185–195. [Google Scholar]
- Shieh, J.; Wu, H.H. Measures of consistency for DEMATEL method. Commun. Stat. Simul. Comput. 2016, 45, 781–790. [Google Scholar] [CrossRef]
- Farhadi, P.; Niyas, M.; Shokrpour, N.; Ravangard, R. Prioritizing Factors Affecting Health Service Quality using Integrated Fuzzy DEMATEL and ANP: A Case of Iran. Open Public Health J. 2020, 13, 263–272. [Google Scholar] [CrossRef]
- García-Cascales, M.S.; Lamata, M.T. On rank reversal and TOPSIS method. Math. Comput. Model. 2012, 56, 123–132. [Google Scholar] [CrossRef]
- Wang, H.; Zheng, H. Model Validation, Machine Learning. In Encyclopedia of Systems Biology; Dubitzky, W., Wolkenhauer, O., Cho, K.H., Yokota, H., Eds.; Springer: New York, NY, USA, 2013. [Google Scholar] [CrossRef]
- Chan, H.P.; Sahiner, B.; Wagner, R.F.; Petrick, N. Classifier design for computer-aided diagnosis: Effects of finite sample size on the mean performance of classical and neural network classifiers. Med. Phys. 1999, 26, 2654–2668. [Google Scholar] [CrossRef]
- Samala, R.K.; Chan, H.P.; Hadjiiski, L.; Helvie, M.A. Risks of feature leakage and sample size dependencies in deep feature extraction for breast mass classification. Med. Phys. 2021, 48, 2827–2837. [Google Scholar] [CrossRef]
- Ghods, A.; Cook, D.J. A survey of deep network techniques all classifiers can adopt. Data Min. Knowl. Discov. 2021, 35, 46–87. [Google Scholar] [CrossRef] [PubMed]
- Zhou, Y.; Liu, Y. Correlation Analysis of Performance Metrics for Classifier. In Decision Making and Soft Computing—Proceedings of the 11th International FLINS Conference; World Scientific: Sidney, Australia, 2014; pp. 487–492. [Google Scholar] [CrossRef]
- Pereira, R.B.; Plastino, A.; Zadrozny, B.; Merschmann LH, C. Correlation analysis of performance measures for multi-label classification. Inf. Processing Manag. 2018, 54, 359–369. [Google Scholar] [CrossRef]
- Doan, T.; Kalita, J. Predicting run time of classification algorithms using meta-learning. Int. J. Mach. Learn. Cybern. 2017, 8, 1929–1943. [Google Scholar] [CrossRef]
- Doshi-Velez, F.; Kim, B. Towards a rigorous science of interpretable machine learning. arXiv 2017, arXiv:1702.08608. [Google Scholar]
- Jakobsen, J.C.; Gluud, C.; Wetterslev, J.; Winkel, P. When and how should multiple imputation be used for handling missing data in randomised clinical trials—A practical guide with flowcharts. BMC Med. Res. Methodol. 2017, 17, 162. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Zhang, Y.; Xiao, L. Stochastic primal-dual coordinate method for regularized empirical risk minimization. J. Mach. Learn. Res. 2017, 18, 1–42. [Google Scholar]
- Lan, G.; Zhou, Y. An optimal randomized incremental gradient method. Math. Program. 2018, 171, 167–215. [Google Scholar] [CrossRef]
- Petkovic, D.; Altman, R.; Wong, M.; Vigil, A. Improving the Explainability of Random Forest Classifier–User Centered Approach. In Pacific Symposium on Biocomputing 2018; World Scientific: Koala Coast, HI, USA, 2018; pp. 204–215. [Google Scholar] [CrossRef] [Green Version]
- Selvaraju, R.R.; Cogswell, M.; Das, A.; Vedantam, R.; Parikh, D.; Batra, D. Grad-CAM: Visual explanations from deep networks via gradient-based localization. Int. J. Comput. Vis. 2020, 128, 336–359. [Google Scholar] [CrossRef] [Green Version]
- de Leon, A.R.; Soo, A.; Williamson, T. Classification with discrete and continuous variables via general mixed-data models. J. Appl. Stat. 2011, 38, 1021–1032. [Google Scholar] [CrossRef]
- Epaillard, E.; Bouguila, N. Hybrid Hidden Markov Model for Mixed Continuous/Continuous and Discrete/Continuous Data Modeling. In Proceedings of the 2015 IEEE 17th International Workshop on Multimedia Signal Processing, Xiamen, China, 19–21 October 2015. [Google Scholar] [CrossRef]
- Shafizadeh-Moghadam, H.; Valavi, R.; Shahabi, H.; Chapi, K.; Shirzadi, A. Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping. J. Environ. Manag. 2018, 217, 1–11. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Netto, A.L.; Salomon, V.A.P.; Barrios, M.A.O. Multi-criteria analysis of green bonds: Hybrid multi-method applications. Sustainability 2021, 13, 512. [Google Scholar] [CrossRef]
Expert | Profession | Experience (Years) | Current Position | Participation in the Design of Reminding Technologies |
---|---|---|---|---|
1 | Biomedical Engineering | >20 | Professor in Biomedical Engineering | Yes |
2 | Systems Engineering | >10 | Lecturer in Data Analytics | Yes |
3 | Systems Engineering | >20 | Senior Lecturer in Ambient Assisted Living | Yes |
4 | Systems Engineering | >10 | Professor of Image Processing | Yes |
5 | Biomedical Engineering | >10 | Assistant Professor in Signal Analysis | Yes |
6 | Industrial Engineering | >20 | Associate Professor | Yes |
7 | Systems Engineering | >10 | Associate Professor | Yes |
Classifier Selection Criterion | Sub-Criteria | Definition |
---|---|---|
Classifier performance (F1) | Predictive ability (SF1) Computational time (SF2) Negative recall (SF3) Positive recall (SF4) Positive predictive value (SF5) Negative predictive value (SF6) | It measures the predictive capability of a classification algorithm; in this context, how well the classifier distinguishes adopters and non-adopters of a particular AT [20]. |
Applicability (F2) | Ease of comprehension by non-experts (SF7) Interpretability (SF8) | This factor denotes how explainable the algorithm is and verifies whether it is easy to understand by clinicians who are often unskilled in this kind of application. This is of interest considering that medical staff will be directly involved in the classifier implementation. |
Replicability (F3) | No sub-criteria | This criterion considers the financial investment underpinning the classifier development process as well as its validation in the practical scenario. |
Adaptability (F4) | Missing data estimation (SF9) Management of discrete and continuous variables (SF10) Online learning (SF11) | It evaluates how flexible the algorithm is when addressing common data drawbacks (i.e. missing data), different implementation conditions, and diverse variable types. Not effectively responding to this context may limit the application of the classifier in the real world. |
Classifier architecture (F5) | Data gathering (SF12) Overtraining effect (SF13) Amount of input data (SF14) Validation (SF15) Statistical classification (SF16) | It exhibits different classifier design aspects including data gleaning, training, and validation which may flatten the learning curve of clinicians while laying the groundwork for the design of agile healthcare processes for PwD. |
Criterion (F)/Sub-Criterion (SF) | D + RT | D − RT | Dispatcher | Receiver |
---|---|---|---|---|
Classifier performance (F1) | 9.595 | −0.371 | X | |
Predictive ability (SF1) | 10.306 | 0.883 | X | |
Computational time (SF2) | 8.582 | −1.141 | X | |
Negative recall (SF3) | 9.425 | −0.289 | X | |
Positive recall (SF4) | 9.711 | 0.230 | X | |
Positive predictive value (SF5) | 9.583 | 0.376 | X | |
Negative predictive vaalue (SF6) | 9.647 | −0.058 | X | |
Applicability (F2) | 9.991 | −0.641 | X | |
Ease of comprehension (SF7) | 43.118 | −1.000 | X | |
Interpretability (SF8) | 43.118 | 1.000 | X | |
Replicability (F3) | 9.075 | −1.333 | X | |
Adaptability (F4) | 9.755 | 1.609 | X | |
Missing data estimation (SF9) | 14.416 | −1.314 | X | |
Management of continuous and discrete variables (SF10) | 14.524 | 0.258 | X | |
Online learning (SF11) | 14.252 | 1.056 | X | |
Classifier architecture (F5) | 10.554 | 0.736 | X | |
Data gathering (SF12) | 12.827 | −0.173 | X | |
Overtraining effect (SF13) | 11.584 | −1.520 | X | |
Amount of input data (SF14) | 12.868 | 0.322 | X | |
Validation (SF15) | 13.039 | 1.174 | X | |
Statistical classification (SF16) | 12.589 | 0.198 | X |
IF-DEMATEL | ||
---|---|---|
Criterion (F)/Sub-Criterion (SF) | LW | GW |
Classifier performance (F1) | 0.196 | |
Predictive ability (SF1) | 0.180 | 0.035 |
Computational time (SF2) | 0.150 | 0.029 |
Negative recall (SF3) | 0.165 | 0.032 |
Positive recall (SF4) | 0.170 | 0.033 |
Positive predictive value (SF5) | 0.167 | 0.033 |
Negative predictive value (SF6) | 0.168 | 0.033 |
Applicability (F2) | 0.204 | |
Ease of comprehension (SF7) | 0.500 | 0.102 |
Interpretability (SF8) | 0.500 | 0.102 |
Replicability (F3) | 0.216 | |
Adaptability (F4) | 0.199 | |
Missing data estimation (SF9) | 0.334 | 0.066 |
Management of continuous and discrete variables (SF10) | 0.336 | 0.067 |
Online learning (SF11) | 0.330 | 0.066 |
Classifier architecture (F5) | 0.216 | |
Data gathering (SF12) | 0.204 | 0.044 |
Overtraining effect (SF13) | 0.184 | 0.040 |
Amount of input data (SF14) | 0.205 | 0.044 |
Validation (SF15) | 0.207 | 0.045 |
Statistical classification (SF16) | 0.200 | 0.043 |
Sub-Factor/Factor | Key Performance İndex | Mathematical Formula |
---|---|---|
Predictive ability (SF1) | Average accuracy | TN: True negative predictions TP: True positive predictions FP: False positive predictions FN: False negative predictions n: Number of iterations |
Computational time (SF2) | Average run time | |
Negative recall (SF3) | Average recall (-) | TN: True negative predictions FP: False positive predictions |
Positive recall (SF4) | Average recall (+) | TP: True positive predictions FN: False negative predictions |
Positive predictive value (SF5) | Average precision (+) | TP: True positive predictions FP: False positive predictions |
Negative predictive value (SF6) | Average precision (-) | TN: True negative predictions FN: False negative predictions |
Ease of comprehension (SF7) | Model appropriation | If the algorithm is easy to appropriate by physicians and clinicians (1); otherwise (0) |
Interpretability (SF8) | Box type | If it a black-box algorithm (0); white-box algorithm (1) |
Replicability (F3) | Unit replication cost | It the learning process cost is higher than £727.48 (0); otherwise (1) |
Missing data estimation (SF9) | Capability of missing data management | If the algorithm is capable of handling missing data (1); otherwise (0) |
Management of continuous and discrete variables (SF10) | Management of continuous and discrete variables | If the algorithm works with both continuous and discrete variables (1); otherwise (0) |
Online learning (SF11) | Online learning | If the algorithm is of online-learning type (1); otherwise (0) |
Data gathering (SF12) | Easiness of data collation | If the feature set of the model can be collated through available data sources and/or simple self-administered surveys (1); otherwise (0) |
Overtraining effect (SF13) | Overtraining | If the algorithm evidences overtraining effect (0); otherwise (1) |
Amount of input data (SF14) | Number of input variables | Number of patient features that the classifier needs for displaying the prediction |
Validation (SF15) | Access to validation datasets | If the algorithm has access to validated datasets (1); otherwise (0) |
Statistical classification (SF16) | Algorithm nature | If the algorithm is based on statistical modelling (1); otherwise (0) |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 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
Ortíz-Barrios, M.A.; Garcia-Constantino, M.; Nugent, C.; Alfaro-Sarmiento, I. A Novel Integration of IF-DEMATEL and TOPSIS for the Classifier Selection Problem in Assistive Technology Adoption for People with Dementia. Int. J. Environ. Res. Public Health 2022, 19, 1133. https://doi.org/10.3390/ijerph19031133
Ortíz-Barrios MA, Garcia-Constantino M, Nugent C, Alfaro-Sarmiento I. A Novel Integration of IF-DEMATEL and TOPSIS for the Classifier Selection Problem in Assistive Technology Adoption for People with Dementia. International Journal of Environmental Research and Public Health. 2022; 19(3):1133. https://doi.org/10.3390/ijerph19031133
Chicago/Turabian StyleOrtíz-Barrios, Miguel Angel, Matias Garcia-Constantino, Chris Nugent, and Isaac Alfaro-Sarmiento. 2022. "A Novel Integration of IF-DEMATEL and TOPSIS for the Classifier Selection Problem in Assistive Technology Adoption for People with Dementia" International Journal of Environmental Research and Public Health 19, no. 3: 1133. https://doi.org/10.3390/ijerph19031133