Most of the open-source decision tree algorithms are based on three splitting criteria (Entropy, Gini Index, and Gain Ratio). Therefore, the advantages and disadvantages of these three popular algorithms need to be studied more thoroughly. Comparisons of the three algorithms were mainly performed with respect to the predictive performance. In this work, we conducted a comparative experiment on the splitting criteria of three decision trees, focusing on their interpretability. Depth, homogeneity, coverage, lift, and stability were used as indicators for measuring interpretability. To measure the stability of decision trees, we present a measure of the stability of the root node and the stability of the dominating rules based on a measure of the similarity of trees. Based on 10 data collected from UCI and Kaggle, we compare the interpretability of DT (Decision Tree) algorithms based on three splitting criteria. The results show that the GR (Gain Ratio) branch-based DT algorithm performs well in terms of lift and homogeneity, while the GINI (Gini Index) and ENT (Entropy) branch-based DT algorithms performs well in terms of coverage. With respect to stability, considering both the similarity of the dominating rule or the similarity of the root node, the DT algorithm according to the ENT splitting criterion shows the best results.

This paper is a follow up to the previous study which reveals that smartphone users are divided into three subcategories according to their usage characteristics. In this paper, these groups are called as ‘general’, ‘entertainment’, and ‘work-assistant’, taking into account their respective characteristics. The ‘general’ is a group whose smartphone usage characteristics are not focused on a specific purpose, the ‘entertainment’ is focused on music, internet, SNS, picture, and e-banking, and the ‘work-assistant’ is on work, GPS, diary. Inter-relation between the importance and satisfaction for the purchase determinants to the groups is investigated. In addition, Kano analysis of quality attributes is also performed, which includes quality type, satisfaction/dissatisfaction index, and PCSI (Potential Customer Satisfaction Improvement) index. The analysis result are as follows. Firstly, inter-relation between importance and satisfaction differs by user group. ‘Internet’, ‘Ease of use’, and ‘Performance’ purchase determinants are evaluated as competitive determinants in ‘work-assistant’ user group. Secondly Kano quality types of quality characteristics also differs by user group. ‘Application’ was classified as an attractive (A) types to ‘entertainment’ group and so on. ‘Internet’ ‘Failure/Bug’, ‘Touch response rate’ and ‘Charging’ are located in ‘Nice’ Region of S-PCSI Diagram and have to be considered as strategic quality characteristics. The results of this study is expected to give some helps in establishing a customer tailored quality strategy.

Since the appearance of smartphones, the smartphone market has been in fierce completion by new technologies and marketing trends. The smartphone market is now somewhat saturated, and the manufacturers are trying to improve their position in the market through the repurchase of existing customers and the influx of competitors. At the same time, customers have their own purchasing criteria for smartphones. Therefore, manufacturers need to determine new technology and marketing trends based on customer purchasing trends and usage characteristics. The aim of this study is to analyze the quality attributes of smartphones. We conducted a survey on 220 respondents, and divided the respondents into several groups by purchasing trends and usage characteristic through cluster analysis. The groups are analyzed and compared based on the Kano model for the quality attributes of smartphone. The analysis result are as follows. Firstly, purchasing trends divide responders into groups that prefers high-end premium smartphones and those that take into account practicality in terms of purchasing trends. Secondly, usage characteristic divide responders into three groups: those with clear usage pattern, those who prefer ease of use, and the rest, and we find out that those with clear usage pattern are important customer in viral marketing. Lastly, Kano analysis is revealed the ‘Slow/ hi-speed camera’, ‘Private mode’, ‘Widget’, ‘Health care’ are attractive quality attributes.