Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Ambidextrous : Journal of Innovation, Efficiency and Technology in Organization
Haikal Yudistira Simarmata
Program Studi Sistem Informasi Fakultas Ilmu Komputer Dan Teknologi Informasi Universitas Muhammadiyah Sumatera Utara
Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Economics, Econometrics & Finance Electrical & Electronics Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Analysis of Changes in Customer Sentiment in the Period Before and After the Google Review-Based Digital Promotion Program as a Basis for Managerial Decision Making at Kopi Kenangan Haikal Yudistira Simarmata
Ambidextrous Journal of Innovation Efficiency and Technology in Organization Vol. 5 No. 03 (2026): Ambidextrous: Journal of Innovation, Efficiency and Technology in Organization
Publisher : Takaza Innovatix Labs Ltd.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61536/ambidextrous.v5i03.893

Abstract

Customer reviews on Google Reviews can be a valuable source of information for understanding consumer response trends toward Kopi Kenangan products and services. However, the large number of reviews and the unstructured text data require computational processing for systematic analysis. This study aims to determine the distribution of customer sentiment before and after the digital promotion program and apply Multinomial Naive Bayes to classify reviews. The study used sentiment analysis and text mining on 500 reviews from 15 Kopi Kenangan outlets in North Sumatra, consisting of 250 reviews before the promotion and 250 reviews after the promotion. Data were obtained from Google Reviews with ratings and review text as analysis instruments. Data underwent preprocessing, TF-IDF unigram and bigram weighting, 80:20 training and testing split, and oversampling of the training data. The results showed an accuracy of 71.00% and a macro F1-score of 61.48%. The distribution of positive sentiment increased from 48.40% to 63.20%, while negative sentiment decreased from 31.60% to 20.40%. In conclusion, there was a change in the distribution of sentiment towards a more positive direction in the period after the promotion, but this change cannot yet be stated as a direct impact of the digital promotion.
Co-Authors