Articles

Implementation Importance-Performance Analysis Method to Increase Customer Satisfaction of Honda Motorcycle Dealer using Text Mining

/in /by

Transportation is an essential aspect of people’s lives that plays an essential role in supporting economic, social, and cultural activities. Private transportation has become a basic need for Indonesian people, increasing yearly, especially for motorcycles. One of the biggest motorcycle brands in Indonesia is Honda, and West Java is one of the most significant contributors to motorcycle usage. However, their market share is decreasing through the years. It is essential to review customer satisfaction in purchasing Honda products both in terms of product and service through their customer journey in purchasing Honda products. Therefore, this research will discover the aspects customers need from online reviews, especially Google reviews. This research conducted case studies on Dealer Daya Adicipta Motora Bandung, Dealer Bintang Niaga Jaya Bogor, and Dealer Murni Motor II Bogor.

Using Importance Performance Analysis (IPA) will recommend aspects that must be prioritized to improve. The first step is to identify the customer’s needs from Google reviews using non-negative matrix factorization (NMF) that produces matrix H and matrix W as the keywords for every aspect. Then, this research also calculates the sentiment for each review using a dependency tree and lexicon SenticNet5. With the output from NMF and sentiment, we will calculate the performance and importance levels. Ultimately, the Performance Level and Importance Level will be plotted into the graph and divided into four quadrants, with Quandrant A as the top priority for improvement. Eight aspects identified in Dealer Daya Adicipta Motora Bandung, the waiting room facility become the priority of improvement. Dealer Bintang Niaga Jaya Bogor has 7 aspects identified, and the services and buying experience become the top priority of improvement; meanwhile, there are 4 aspects identified in Dealer Murni Motor II Bogor, with the waiting time and waiting room facility become the top priority for improvement.

Perceptive Influence of Purchasing Motor Vehicle Insurance Policy from Non-regulated Firms on the Performance of Insurance Industry in Nigeria: A Customer-Based Sentiment Analysis

/in /by

This study examined Customers’ perception on the influence of purchasing motor vehicle insurance policy from non-regulated firms on the performance of insurance industry in Nigeria. Specifically, the influence of purchasing fake insurance policy and non-renewal of expired policy bought from the regulated insurance firms on the performance of insurance industry in Nigeria were accessed.  Primary data collected through the use of structured questionnaire from 92 vehicle owners in Uyo, Akwa Ibom State, Nigeria, that were selected through convenience sampling technique, was used in the study. Sentiment analysis was applied as the method of data analysis in the study. The result of the analysis indicated a neutral sentiment level of the respondents to both the influence of purchasing fake motor vehicle insurance policy and impact of non-renewal of expired motor vehicle insurance policy bought from regulated insurance firms on the performance of insurance industry in Nigeria. These findings implied a neutral perception of the effect of purchasing motor vehicle insurance policy from non-regulated insurance firms on the performance of insurance industry in Nigeria by the motor vehicle owners. The neutral perception found explains the below expectation performance of the insurance industry in Nigeria due to poor patronage. To enhance public perception in the insurance industry in Nigeria, targeted product features and benefits awareness campaigns as well as payment of genuine claims to deepen public trust in the industry were recommended.

Predicting Customer Satisfaction through Sentiment Analysis on Online Review

/in /by

User-generated content, such as user reviews, posts, tags, ratings, and opinions on the internet, can be used as a business indicator if collected and appropriately analyzed. One of the examples is predicting customer satisfaction through implementing big data analytics on online reviews. In analyzing the user-generated content to predict customer satisfaction, the author implements machine learning approach using the Sentiment Analysis method. Five-fold cross-validation was performed to train the classification model. The training was performed with a combination of tokenization methods: term frequency-inverse document frequency (tf-idf) and bag-of-words; n-gram types: unigram, bigram, trigram, and combination of unigram, bigram, and trigram; and machine learning algorithms: linear support vector classification (LinearSVC) and multinomial naïve bayes (MultinomialNB). The result was then evaluated using classification performance metrics such as precision, recall, F1 measure, and AUC score.

The result shows that the tf-idf vectorizer performs similarly to the bag-of-words method. A similar result was also observed for machine learning algorithm selection. Both MultinomialNB and LinearSVC produce the same performance. Low-level n-grams (such as unigrams and bigrams) tended to have higher precision, recall, F1 measure, and AUC score than high-order n-grams (such as trigrams). The best results were achieved by combining unigrams, bigrams, and trigrams, resulting in an average performance score of 0.94 for all measurements. From the result and analysis, the author finds that predicting customer satisfaction using text and sentiment analysis methods on user-generated content is possible. The model’s performance in this experiment is decent, with high precision, recall, F1, and AUC score.