Extraction of attribute importance from satisfaction surveys with data mining techniques: a comparison between neural networks and decision trees

Oña, Juan de; Oña, Rocío de; Garrido, Concepción Domínguez

doi:10.1080/19427867.2015.1136917

Cited by 14 publications

(9 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, a careful view of Table 1 reveals that very few studies have focused on inter-city transportation services. Moreover, very few studies actually compare the efficacy of different modeling techniques in order to find the optimum model in the domain of SQ studies [24]- [25], [36]. Also, it is noticeable that despite the use of machine learning models like decision trees in different SQ studies, researchers are yet to use any ensemble models like Random Forests or more fine-tuned applications of decision trees like gradient boosted decision tree models.…”

Section: Discussion and Future Research Directionsmentioning

confidence: 99%

“…However, decision tree models are not typically robust, have lower accuracy rates, do not offer backtracking techniques, or provide statistical significance of variables [36]. Moreover, studies using decision trees have not specifically mentioned the algorithms used [13], [21].…”

Section: Decision Treesmentioning

confidence: 99%

“…A number of advantages of using neural networks have been reported by researchers higher precision, more stability for determining attribute importance, capability for handling multicollinearity [36]- [37]. The main drawback of using neural networks reported is mainly related to the time necessary to train and the cumbersome routine needed to determine variable importance.…”

Section: Neural Networkmentioning

confidence: 99%

“…The different types of neural networks employed in SQ studies and often compared to each other or other techniques include ANN [37], comparison between ANN and DT [36] comparison between ANFIS and PNN [24], comparison between PNN, PRNN, and GRNN [25] and ANFIS [22].…”

Section: Neural Networkmentioning

confidence: 99%

See 3 more Smart Citations

A Review of Methodological Approaches and Modeling Techniques in Service Quality Evaluation of Surface Transportation during the Last Decade

Islam

2021

J. eng. adv.

View full text Add to dashboard Cite

During the duration of the last decade, a growing interest has been noticed among transport practitioners and researchers to better understand the concept of service quality in the field of surface transportation and identify important service quality (SQ) attributes of different transportation services since these results have implications for transport managers. Due to advancements in computer technology and the availability of software packages, researchers are better able to extract meaningful results from passengers’ opinions collected through stated preference surveys and communicate their findings to transport managers looking to ameliorate SQ to boost ridership on a limited budget. Since the concept of SQ is itself complex owing to the nature of the service itself compared to a tangible product and characteristics of SQ attribute, different advanced modelling techniques based on multivariate analysis, machine learning, and artificial intelligence paradigms have become popular tools among researchers. This paper aims to summarize the trends of the SQ research in the field of surface transportation during the last decade with a focus on the methodological approaches and modelling techniques and delineate future directions for research in this field.

show abstract

Section: Discussion and Future Research Directionsmentioning

confidence: 99%

Section: Decision Treesmentioning

confidence: 99%

Section: Neural Networkmentioning

confidence: 99%

Section: Neural Networkmentioning

confidence: 99%

See 2 more Smart Citations

A Review of Methodological Approaches and Modeling Techniques in Service Quality Evaluation of Surface Transportation during the Last Decade

Islam

2021

J. eng. adv.

View full text Add to dashboard Cite

show abstract

“…These include the use of decision trees [17][18][19][20][21], multi-criteria analysis [22,23], factor analysis [24][25][26][27], neural network methods [28][29][30] or multiple regression methods [14,31,32].…”

Section: Introductionmentioning

confidence: 99%

Weighted Variables Using Best-Worst Scaling in Ordered Logit Models for Public Transit Satisfaction

et al. 2020

View full text Add to dashboard Cite

Customer overall satisfaction regarding a public transport system is dependent on the satisfaction of the users with the attributes that make up the service, as well as the contribution that each of these attributes makes to explain the overall satisfaction. A common way of analysing the contribution of service attributes to explain overall satisfaction is through the use of ordered logit or probit models. This article presents an ordered logit model that considers the weighting of independent variables through the explicit importance calculated on the basis of a best-worst case 1 choice task. For the calculation of importance, a multinomial logit model has been estimated which considers the heterogeneity of the sample through systematic variations in user tastes. In this way, it is possible to establish a level of importance of each specific attribute for each type of user. The results show that the importance varies considerably depending on different socio-economic and mobility-base variables. On the other hand, the inclusion of the weighted variables in the ordered logit model improves its fit. Therefore, the results make possible to develop policies focused on improving satisfaction on specific user targets.

show abstract

Transit service quality analysis using cluster analysis and decision trees: a step forward to personalized marketing in public transportation

2015

View full text Add to dashboard Cite

A transit service quality study based on cluster analysis was performed to extract detailed customer profiles sharing similar appraisals concerning the service. This approach made it possible to detect specific requirements and needs regarding the quality of service and to personalize the marketing strategy. Data from various customer satisfaction surveys conducted by the Transport Consortium of Granada (Spain) were analyzed to distinguish these groups; a decision tree methodology was used to identify the most important service quality attributes influencing passengers' overall evaluations. Cluster analysis identified four groups of passengers. Comparisons using decision trees among the overall sample of all users and the different groups of passengers identified by cluster analysis led to the discovery of differences in the key attributes encompassed by perceived quality.

show abstract

Extraction of attribute importance from satisfaction surveys with data mining techniques: a comparison between neural networks and decision trees

Cited by 14 publications

References 30 publications

A Review of Methodological Approaches and Modeling Techniques in Service Quality Evaluation of Surface Transportation during the Last Decade

A Review of Methodological Approaches and Modeling Techniques in Service Quality Evaluation of Surface Transportation during the Last Decade

Weighted Variables Using Best-Worst Scaling in Ordered Logit Models for Public Transit Satisfaction

Transit service quality analysis using cluster analysis and decision trees: a step forward to personalized marketing in public transportation

Contact Info

Product

Resources

About