Claim Missing Document
Check
Articles

Found 11 Documents
Search

Sistem Rekomendasi Skincare Berdasarkan Jenis Kulit Menggunakan Content-Based Filtering dan Knowledge-Based Normalization Joy Raphaela; Arif Nur Rohman; Ika Nur Fajri
METIK Jurnal Vol. 10 No. 1 (2026): METIK Jurnal Issue Published
Publisher : LPPM Universitas Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47002/aa1yds57

Abstract

The rapid growth of the skincare industry has triggered information overload, complicating consumer decision-making particularly among Generation Z users on e-commerce platforms. Conventional Collaborative Filtering approaches are limited by popularity bias and the cold-start problem, and are unable to account for ingredient-level compatibility with individual skin conditions. Addressing this gap, this study proposes a novel Content-Based Filtering recommendation system that integrates TF-IDF and Cosine Similarity with a Knowledge-Based Normalization layer. This original framework maps informal consumer terminology into standardized dermatological categories, effectively reducing semantic inconsistency in unstructured product descriptions. Data were obtained from the Kaggle public repository (third-party extracted dataset) and underwent a validation process, yielding a final dataset of 91 skincare products. The system was evaluated using Precision@K across five skin-condition scenarios. Results yield an average Precision@5 of 0.80 (80%), with a peak cosine similarity score of 0.3606. The low absolute cosine value is attributable to TF-IDF vector sparsity in short-text descriptions, a characteristic acknowledged in prior literature. Implementation as a web application confirms the system's practical utility in guiding users toward biologically appropriate skincare choices, independent of market-trend bias.
PENERAPAN ALGORITMA NAÏVE BAYES DENGAN TEKNIK SMOTE UNTUK KLASIFIKASI SENTIMEN KURSUS ONLINE SKILL ACADEMY: APPLICATION OF THE NAÏVE BAYES ALGORITHM WITH SMOTE TECHNIQUE FOR SENTIMENT CLASSIFICATION OF SKILL ACADEMY ONLINE COURSES Arsellina Milka Martin; Arif Nur Rohman
Rabit : Jurnal Teknologi dan Sistem Informasi Univrab Vol 11 No 1 (2026): Januari
Publisher : LPPM Universitas Abdurrab

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36341/rabit.v11i1.7545

Abstract

The rapid growth of online learning has led to the emergence of various e-learning platforms, including Skill Academy. However, not all courses are able to maintain learner engagement, partly due to discrepancies between user expectations and the quality of the provided materials. This study aims to classify user sentiment toward course titles by applying the Multinomial Naïve Bayes algorithm combined with the Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance. Data were collected through web scraping from five main public course pages, including information on course titles, prices, ratings, number of raters, release dates, and topic categories. Sentiment labels were assigned based on rating values, where ratings ≥ 4.0 were categorized as positive and ratings < 4.0 as negative. Text feature extraction was performed using the TF-IDF method. The experimental results show that the model developed without SMOTE achieved an accuracy of 89.36% but completely failed to identify the negative class, as indicated by a recall value of 0%. After applying SMOTE to the training data, the recall for the negative class increased significantly to 64% demonstrating a substantial improvement in the model’s ability to detect previously overlooked negative sentiment. Although a slight decrease in accuracy was observed in several testing scenarios, the improvement in recall and F1-score for the minority class represents the primary contribution of this study. These findings confirm that SMOTE is effective in mitigating class imbalance and enhances sentiment analysis performance for short text data on online course platforms.
Implementation of the Apriori Algorithm for Clothing Store Product Recommendations based on Sales Transaction History M Ilham Saputro; Arif Nur Rohman
SISTEMASI Vol 15, No 1 (2026): Sistemasi: Jurnal Sistem Informasi
Publisher : Universitas Islam Indragiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v15i1.5648

Abstract

This study is motivated by the limitations faced by small-scale clothing stores, which generally do not have customer ratings or reviews that can be used as a basis for product recommendations. This condition necessitates an alternative method capable of utilizing available sales transaction data. The objective of this study is to generate product recommendations by identifying consumer purchasing patterns through the application of the Apriori Algorithm. The methodology involves processing sales transaction data consisting of transaction codes, lists of purchased products, and transaction timestamps. Support, confidence, and lift ratio values are calculated to generate and validate association rules among products. The analyzed data are derived from the transaction history of a clothing store and are processed using a web-based system developed with PHP and MySQL. The experimental results indicate that several product combinations achieve confidence values of 50% and lift ratios greater than or equal to 1, suggesting that these patterns can be used as a basis for product recommendations. These findings demonstrate a strong association among items that are frequently purchased together. Based on the results, this study concludes that the Apriori Algorithm is effective in identifying meaningful purchasing patterns that can support product arrangement strategies and inventory management in small-scale clothing stores.
Aceh Province Tourism Destination Recommendation System using Content based Filtering Method Ariefhan Maulana; Arif Nur Rohman; Yoga Pristyanto
SISTEMASI Vol 15, No 2 (2026): Sistemasi: Jurnal Sistem Informasi
Publisher : Universitas Islam Indragiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v15i2.5500

Abstract

Tourists often experience difficulties in finding tourist destinations in Aceh Province that match their content preferences and are geographically close to their location. This study aims to develop a tourism destination recommendation system in Aceh Province using a Content-Based Filtering approach with the Cosine Similarity algorithm and the Haversine Formula. The dataset consists of 119 tourist destinations, including attributes such as destination name, destination description, and geographical coordinates (latitude and longitude). The research process began with text data preprocessing, which included case folding, punctuation removal, tokenization, duplicate word removal, stopword removal, and stemming. Next, the similarity between destinations was calculated using the Cosine Similarity algorithm based on tourism content descriptions, while the Haversine Formula was applied to measure the geographical distance between the user’s location and the tourist destinations. The results indicate that the developed system is able to provide relevant tourism destination recommendations by simultaneously considering content relevance and geographical proximity. Therefore, the system can assist tourists in selecting destinations that best match their preferences.
Implementation of a Hybrid Filtering Approach in a Website-based Football News Recommendation System Krissna Haridarma Putra; Arif Nur Rohman; Norhikmah hikmah
SISTEMASI Vol 15, No 1 (2026): Sistemasi: Jurnal Sistem Informasi
Publisher : Universitas Islam Indragiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v15i1.5854

Abstract

The rapid growth of football news on digital portals has made it increasingly difficult for users to find information that matches their interests. This study develops a web-based news recommendation system by combining Content-Based Filtering and Collaborative Filtering through a feature-level Hybrid Filtering approach. The proposed hybrid approach constitutes the main novelty of this research, as it does not rely on score aggregation methods commonly used in previous studies, making it lighter, simpler, and more suitable for small datasets and limited user interactions. The system employs Term Overlap Matching to measure the similarity between news titles and Cosine Similarity to assess user preference similarity based on bookmark data. The evaluation results show that Content-Based Filtering achieves the best performance, with a Precision of 0.60, Recall of 0.75, and an F1-score of 0.67, while Collaborative Filtering performs poorly due to data sparsity in user interactions, resulting in a Precision of 0, Recall of 0, and an F1-score of 0. Overall, the feature-based hybrid approach is able to provide relevant recommendations from both content and preference perspectives, although system accuracy is still predominantly driven by Content-Based Filtering. These findings indicate that the proposed simple hybrid model can serve as an effective solution for small-scale sports news platforms and has the potential to be further improved through increased data availability, enhanced user interaction, and the adoption of more advanced NLP techniques.
PENERAPAN METODE SIMPLE ADDITIVE WEIGHTING DALAM SISTEM PENDUKUNG KEPUTUSAN PEMILIHAN TEMPAT KOS DI CONDONG CATUR YOGYAKARTA Ridhwan Shodiq Ramadhan; Arif Nur Rohman; Alfie Nur Rahmi
Information System Journal Vol. 8 No. 02 (2025): Information System Journal (INFOS)
Publisher : Universitas Amikom Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24076/infosjournal.2025v8i02.2429

Abstract

Pemilihan tempat kos yang ideal merupakan tantangan besar bagi mahasiswa yang ingin melanjutkan pendidikannya keluar kota. Namun, proses pencarian proses penilaian fasilitas tempat kos sering kali masih harus dilakukan secara langsung, yang menghabiskan waktu dan biaya. Oleh karena itu dibangunlah sistem pendukung keputusan menggunakan metode Simple Additive Weighting (SAW) berbasis website yang dapat membantu mahasiswa dalam menentukan tempat kos sesuai dengan kriteria yang diinginkan. Metode penelitian dilakukan melalui beberapa tahap: pengumpulan data dari sumber terkait, analisis sistem untuk menyusun alur kerja, perancangan sistem rekomendasi, dan implementasi sistem. Berdasarkan hasil perhitungan perangkingan, Residence Permai terpilih sebagai rekomendasi terbaik dengan nilai akhir 0.8500, diikuti oleh Griya Seturan dengan nilai 0,8400, dan Kos Candi Gebang dengan nilai 0,8000. Hasil pengujian menunjukkan bahwa sistem ini berjalan sesuai dengan yang diharapkan dan mampu memberikan solusi yang lebih komprehensif dan praktis bagi mahasiswa yang mencari tempat kos di Condong Catur, Sleman, Yogyakarta
Application of Content-Based Filtering Method Using Cosine Similarity in Restaurant Selection Recommendation System Fajar Christyawan; Arif Nur Rohman; Anggit Dwi Hartanto
Journal of Information System and Informatics Vol 6 No 3 (2024): September
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51519/journalisi.v6i3.806

Abstract

This research focuses on developing a restaurant recommender system designed to assist users in selecting restaurants based on preferences such as cuisine type and proximity, thereby enhancing the dining experience. The system employs a content-based filtering approach combined with the Cosine Similarity algorithm to calculate similarity values between restaurant addresses and categories, ensuring personalized and accurate recommendations. Data for the system was collected from TripAdvisor and Google Maps using a web scraping method, resulting in a comprehensive dataset that reflects a wide variety of dining options. An experiment involving 30 respondents was conducted to evaluate the system's performance under real-world conditions. The results demonstrated an accuracy rate of 88%, indicating that the recommender system effectively delivers highly relevant restaurant suggestions to users. These findings suggest that the system can serve as a valuable tool for culinary tourists and local residents, simplifying the process of discovering new dining experiences and aligning them with individual preferences.
Hotel Recommendation System in Yogyakarta based on Content-Based Filtering with FastText Modeling Dzikri Albantani; Arif Nur Rohman
SISTEMASI Vol 15, No 10 (2026): Sistemasi: Jurnal Sistem Informasi
Publisher : Universitas Islam Indragiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v15i10.6336

Abstract

The rapid growth of tourism in Yogyakarta has resulted in a wide range of accommodation options, making it challenging for tourists to find hotels that meet their needs. Recommendation systems based on Term Frequency–Inverse Document Frequency (TF-IDF) have limitations in handling variations in word forms and informal language because their matching process relies on lexical representations. This study aims to develop a hotel recommendation system using Content-Based Filtering integrated with FastText word embeddings and to compare its performance with TF-IDF in handling variations between formal and informal language. The study uses a dataset of 100 hotels in Yogyakarta collected through web scraping and data validation. The evaluation was conducted using five pairs of formal and informal queries with equivalent search intents. For each query and method, the top five recommendations were used to calculate Mean Similarity@5 and measure score changes from formal to informal queries. The experimental results show that changes in Mean Similarity@5 for FastText tended to be smaller than those for TF-IDF across the five query pairs tested. The largest change observed for TF-IDF was 26.09%, compared with 5.41% for FastText. These findings indicate that FastText is relatively more robust to variations between formal and informal language under the evaluation scenarios considered in this study. However, the study does not use independently established ground-truth relevance labels; therefore, the evaluation results are not used to claim differences in recommendation accuracy. The analysis also reveals limitations in handling numerical attributes, such as price, and certain syntactic relationships, such as negation.
A Tourist Village Recommendation System in Sleman using Content-based Filtering, TF-IDF, and Cosine Similarity Herda Rafif Arya Putra; Arif Nur Rohman
SISTEMASI Vol 15, No 10 (2026): Sistemasi: Jurnal Sistem Informasi
Publisher : Universitas Islam Indragiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v15i10.6824

Abstract

Sleman Regency, located in the Special Region of Yogyakarta, is home to numerous tourist villages offering diverse attractions, ranging from natural and cultural tourism to culinary experiences. However, many tourists still face difficulties in finding tourist villages that match their interests and needs because the available information tends to be general rather than personalized. This study develops a tourist village recommendation system for Sleman using a Content-Based Filtering approach combined with Term Frequency–Inverse Document Frequency (TF-IDF) and cosine similarity. The dataset comprises 20 active tourist villages in Sleman Regency, compiled from the official website of the Sleman Regency Tourism Office in 2024. The dataset includes the name of each tourist village, subdistrict, tourism category, main attractions, and facilities. These attributes were combined into a single descriptive text for each village and processed through case folding, tokenization, stopword removal, and stemming using the Sastrawi library, resulting in 125 unique features. The resulting text representations were then weighted using TF-IDF. Cosine similarity was applied to measure the similarity between user preference queries and each tourist village, and the five villages with the highest similarity scores were presented as recommendations (Top-5). The system was evaluated using three query scenarios and the Precision@5, Recall@5, and F1-score metrics, with ground-truth relevance labels manually established based on attribute matching. The evaluation results yielded an average score of 93.3% across Precision@5, Recall@5, and F1-score. This study is expected to help tourists identify suitable tourist villages while supporting the promotion of village tourism potential in Sleman Regency.
Comparative Performance of SVM and Multinomial Naïve Bayes in Sentiment Analysis of the Film 'Dirty Vote' Aisha Shakila Iedwan; Nia Mauliza; Yoga Pristyanto; Anggit Dwi Hartanto; Arif Nur Rohman
Scientific Journal of Informatics Vol. 11 No. 3: Special Issue
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i3.10290

Abstract

Purpose: The purpose of this research is to analyze and compare the performance of two machine learning models, Support Vector Machine (SVM) and Multinomial Naive Bayes, in conducting sentiment analysis on YouTube comments related to the film "Dirty Vote." Methods: The study involved collecting YouTube comments and preprocessing the data through cleaning, labeling, and feature extraction using TF-IDF. The dataset was then divided into training and testing sets in an 80:20 ratio. Both the SVM and Multinomial Naive Bayes models were trained and tested, with their performance evaluated using accuracy, precision, recall, and F1-score metrics. Result: The results revealed that both models performed well in classifying sentiments, with SVM slightly outperforming Multinomial Naive Bayes in terms of accuracy and precision. Particularly, SVM showed superior performance in detecting positive comments, making it a more reliable model for this specific sentiment analysis task. Novelty: This study contributes to the field of sentiment analysis by providing a detailed comparative analysis of SVM and Multinomial Naive Bayes models on YouTube comments in the context of an Indonesian film. The findings highlight the strengths and weaknesses of each model, offering insights into their applicability for sentiment analysis tasks, particularly in analyzing social media content. This research also suggests potential future directions, including the exploration of advanced NLP techniques and different models to enhance sentiment analysis performance.