A Machine Learning Approach to Customer Needs Analysis for Product Ecosystems

Abstract

Creating product ecosystems has been one of the strategic ways to enhance user experience and business advantages. Among many, customer needs analysis for product ecosystems is one of the most challenging tasks in creating a successful product ecosystem from both the perspectives of marketing research and product development. In this paper, we propose a machine-learning approach to customer needs analysis for product ecosystems by examining a large amount of online user-generated product reviews within a product ecosystem. First, we filtered out uninformative reviews from the informative reviews using a fastText technique. Then, we extract a variety of topics with regard to customer needs using a topic modeling technique named latent Dirichlet allocation. In addition, we applied a rule-based sentiment analysis method to predict not only the sentiment of the reviews but also their sentiment intensity values. Finally, we categorized customer needs related to different topics extracted using an analytic Kano model based on the dissatisfaction-satisfaction pair from the sentiment analysis. A case example of the Amazon product ecosystem was used to illustrate the potential and feasibility of the proposed method.

Analysis of Customer Sentiment on Product Features After the Outbreak of Coronavirus Disease (Covid-19) Based on Online Reviews

“…One is to extract product features that the customer considers important from customer reviews and the other is to analyze sentiment indexes such as adjectives related to feature words or the rating of reviews containing feature words. Zhou et al (2020) adopted latent Dirichlet allocation (LDA) for analyzing the review data. LDA returned the topic mentioned in reviews and those that are related to product features were selected.…”

An Automated Method to Conduct Importance-Performance Analysis of Product Attributes From Online Reviews - An Extension With a Case Study

“…These labels will be our product attributes. In the past, LDA has shown success in identifying product attributes from online reviews (Jeong et al, 2019;Zhou et al, 2019;Joung and Kim, 2020).…”

Emotional Design

“…The sample sizes in these Kansei studies were relatively small (< 20) and in order to reduce the possible subjective biases (Pryzant et al, 2020), online product reviews can be readily collected from websites (e.g., Amazon.com) in large quantities. For example, a large amount of review data for Kindle tablets and Amazon product ecosystems were crawled from Amazon to understand reviewers' emotional responses and satisfaction (Zhou et al, 2015;Ayoub, Zhou, Xu, & Yang, 2019;Zhou, Ayoub, Xu, & Jessie Yang, 2020). Human agents or avatars are also used to elicit emotional responses for interactive interfaces.…”

“…In addition, human-computer integration is emerging, in which computational and human systems can be interwoven closely in a wider social-technical system (Mueller et al, 2020). Furthermore, user-generated data on websites, such as online product reviews on Amazon.com, can be utilized to understand and update customer needs more efficiently and effectively with a large number of users on a daily basis (Li, Tian, Wang, Wang, & Huang, 2018;Zhou, Ayoub, et al, 2020). Third, high-performance computing resources, such as graphic processing units and tensor processing units, allow the training of large-scale deep learning models for big data possible (Q.…”