Urrochman, Maysas Yafi
Unknown Affiliation

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Journal of Informatics Development

Sentiment Analysis of Ijen Crater Reviews using Decision Tree Classification and Oversampling Optimization Hizham, Fadhel Akhmad; Asyari, Hasyim; Urrochman, Maysas Yafi
Journal of Informatics Development Vol. 3 No. 1 (2024): Oktober 2024
Publisher : Institut Teknologi dan Bisnis Widya Gama Lumajang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30741/jid.v3i1.1399

Abstract

Sentiment analysis is a text mining technique that classifies content as positive, negative, or neutral polarity in each sentence or document. These lines or papers may be user reviews assessing the quality of a product or material supplied to them. The purpose of this study is to better understand the function of sentiment analysis in assessing evaluations of the Ijen Crater tourist destination based on Google Maps user comments. This study is conducted in four steps, beginning with data gathering in the form of Google Maps evaluations obtained by data scraping. Following data collection, text preparation includes case folding, tokenization, stopword elimination, and stemming. Following text preprocessing, the next stage is imbalaced data optimization, which involves modifying the minority class samples to be nearly equal to the majority class by randomly duplicating minority class samples. Then, each review is categorized according to sentiment using the Decision Tree (DT) method. Testing has done by comparing DT without optimization and DT with SMOTE-ENN and ADASYN optimization. The result shown DT with SMOTE-ENN optimization has the best accuracy improvement with 1.62%, from 96.94% to 98.56%.
AnalysisSentimentAlun-Alun LumajangReviewusingSupportVector Machine Urrochman, Maysas Yafi
Journal of Informatics Development Vol. 3 No. 2 (2025): April 2025
Publisher : Institut Teknologi dan Bisnis Widya Gama Lumajang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30741/jid.v3i2.1555

Abstract

Alun-Alun Lumajangis one of room the public that becomes center activity community and tourists . Perception public to place Thiscan measured through analysis sentiment to reviews available on digital platforms such as Google Maps. Research This aiming For classifysentiment review the use Support Vector Machine (SVM) method , one of the effective machine learning algorithms Fortask classification text . Data used in the form of review collected text fromGoogle Maps, then through pre-processing data such as cleaning text , tokenization , and deletion stopword . Sentiment label determined manually to be three categories : positive, negative , and neutral . Next , the data is extracted use TF-IDF technique before classified using SVM. Research results showthat SVM algorithm is capable of classify sentiment with level high accuracy , making it proper method For analysis opinion public based on text . Findings This expected can give input for government area in increase quality services and management room public in Lumajang.
Aspect-Based Sentiment Analysis of Tumpak Sewu Waterfall Tourist Reviews Using the Naive Bayes Classifier (NBC) Method Urrochman, Maysas Yafi; Asy’ari, Hasyim; Ro’uf, Abdur
Journal of Informatics Development Vol. 4 No. 1 (2025): Oktober 2025
Publisher : Institut Teknologi dan Bisnis Widya Gama Lumajang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30741/jid.v4i1.1758

Abstract

With the increasing popularity of Tumpak Sewu Waterfall, the volume of visitor reviews on Google Maps continues to grow. These reviews contain valuable insights into tourists’ experiences; however, conducting an in-depth manual analysis is inefficient. This study aims to perform aspect-based sentiment analysis on visitor reviews of Tumpak Sewu Waterfall using the Naive Bayes Classifier (NBC) method. This approach enables the classification of sentiments positive, negative, and neutral based on specific aspects such as facilities, accessibility, and natural scenery. Review data were collected from online platforms and processed through stages of text preprocessing and feature extraction before being trained using the NBC model. The results show that the model effectively classifies review sentiments with a high level of accuracy and provides detailed insights into which aspects most influence visitor satisfaction. These findings not only demonstrate the effectiveness of the Naive Bayes Classifier in aspect-based sentiment analysis tasks but also offer data-driven strategic recommendations for tourism managers to enhance service quality and improve visitor experience in the future.