Background: E-commerce websites have been established expressly as useful online communication platforms, which is rather significant. Through them, users can easily perform online transactions such as shopping or ordering food and sharing their experiences or feedback. Objectives: Customers’ views and sentiments are also analyzed by businesses to assess consumer behavior or a point of view on certain products or services. Methods/Approach: This research proposes a method to extract customers’ opinions and analyse sentiment based on a collected dataset, including 236,867 online Vietnamese reviews published from 2011 to 2020 on foody.vn and diadiemanuong.com. Then, machine learning models were applied and assessed to choose the optimal model. Results: The proposed approach has an accuracy of up to 91.5 percent, according to experimental study findings. Conclusions: The research results can help enterprise managers and service providers get insight into customers’ satisfaction with their products or services and understand their feelings so that they can make adjustments and correct business decisions. It also helps food e-commerce managers ensure a better e-commerce service design and delivery.

No abstract

No abstract

On social networks, each message has many features where the interested topics and the actors sending and receiving topics are important features. Unlike the traditional approach, which views each message belonging to a topic, the topic model is based on the approach, which indicates that each message has a mixture of many topics. However, topic model has limitations about discovering interested topics of actors with temporal factor and labelling latent topics. The article proposes a temporal-author-recipient-topic (TART) model based on: (i) discovering interested topics and analyzing the role of actors on social networks with the temporal factor; (ii) labelling the latent topics from topic model based on topic taxonomy; (iii) applying the temporal factor for finding the relation among factors in model; and (iv) finding out the variation of interested topics of actors with each period of time. An experimenting TART model on two corpora with 1,004,396 messages in Vietnamese and 25,009 actors by the software is built for SNA.