cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Andy Sapta
Contact Email
saptaandy@gmail.com
Phone
-
Journal Mail Official
lppm_stmik@royal.ac.id
Editorial Address
-
Location
,
INDONESIA
JOURNAL OF SCIENCE AND SOCIAL RESEARCH
Published by Smart Education
ISSN : 26154307     EISSN : 26153262     DOI : -
Core Subject : Economy, Science, Education, Social,
Computer Science & IT Economics, Econometrics & Finance Education Social Sciences
Journal of Science and Social Research is accepts research works from academicians in their respective expertise of studies. Journal of Science and Social Research is platform to disclose the research abilities and promote quality and excellence of young researchers and experienced thoughts towards Change for Development. The journal releases on February and July.
Arjuna Subject : -
Articles 3,645 Documents
 325   326   327   328   329 
KLASIFIKASI TIPE KONSUMEN BERDASARKAN RIWAYAT TRANSAKSI MENGGUNAKAN METODE K-NEAREST NEIGHBOR (KNN) Al Izzati Karimah; Wanayumini
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 2 (2026): April 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i2.6262

Abstract

Abstract: Advances in digital technology have generated consumer transaction history data that can be used to analyze consumer behavior and types. However, this data has not yet been optimally utilized in the consumer segmentation process. This study aims to develop a consumer type classification system based on transaction history using the K-Nearest Neighbor (KNN) method. The dataset used was sourced from Kaggle and included variables such as product type, brand, category, quantity, price, and payment method. The research stages included data collection, preprocessing, model training and testing, and evaluation of classification results. The system was built using PHP and MySQL. The KNN method was used to group consumers into Regular, Premium, and Optimal categories. Testing results using 100 data points and setting K=5 showed an accuracy of 65%; the study found that the K-Nearest Neighbor (KNN) method successfully classified the data. Keywords: Classification, Consumer Type, Transaction History, Data Mining, K-Nearest Neighbor (KNN). Abstrak: Perkembangan teknologi digital menghasilkan data riwayat transaksi konsumen yang dapat dimanfaatkan untuk menganalisis perilaku dan tipe konsumen. Namun, data tersebut belum dimanfaatkan secara optimal dalam proses pengelompokan konsumen. Penelitian ini bertujuan membangun sistem klasifikasi tipe konsumen berdasarkan riwayat transaksi menggunakan metode K-Nearest Neighbor (KNN). Dataset yang digunakan berasal dari Kaggle dengan variabel jenis produk, merek, kategori, kuantitas, harga, dan metode pembayaran. Tahapan penelitian meliputi pengumpulan data, prapemrosesan, pelatihan dan pengujian model, serta evaluasi hasil klasifikasi. Sistem dibangun menggunakan PHP dan MySQL. Metode KNN digunakan untuk mengelompokkan konsumen ke dalam kategori Reguler, Premium, dan Optimal. Hasil pengujian menggunakan 100 data dan mencari K=5 menunjukkan hasil akurasi sebanyak 65%, dari penelitian yang dilakuakn Metode K-Nearest Neighbor (KNN) berhasil mengklasifikasikan dengan baik. Kata Kunci: Klasifikasi, Tipe Konsumen, Riwayat Transaksi, Data Mining, K-Nearest Neighbor (KNN).
ANALISIS KEPUTUSAN DALAM PENENTUAN MENU FAVORIT PADA PROGRAM MAKAN BERGIZI BERDASARKAN PREFERENSI RASA DAN KANDUNGAN GIZI MENGGUNAKAN METODE FUZZY MAMDANI Fahrizal; Helmi Fauzi Siregar
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 2 (2026): April 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i2.6263

Abstract

Abstract: The success of a nutritious meal program for school children depends heavily on the balance between objective fulfillment of nutritional standards and subjective taste acceptability. Mistakes in menu preparation often result in high food waste due to unpopular taste, despite its high nutritional content. This study aims to conduct a scientific decision analysis in evaluating and determining the feasibility of a food menu through a web-based system interface. The evaluation was analyzed using the Mamdani fuzzy logic algorithm by processing two main input variables, namely the composite nutritional score and the hedonic score. The computational stages of the analysis include fuzzification, MIN implication rule inference, MAX curve aggregation, and defuzzification using the Center of Area (CoA) method based on the area-area continuous integral. The results of the functional testing showed that all computational modules operated 99% validly. This decision analysis successfully determined the final feasibility percentage and classified the menu into three statuses: Recommended, Considered, and Rejected, thus facilitating the determination of favorite menus in a measurable and objective manner. Keywords: Decision Analysis, Fuzzy Mamdani, Nutritious Meal, Hedonic Score, Nutritional Adequacy Rate.   Abstrak: Keberhasilan program makan bergizi bagi anak sekolah sangat bergantung pada keseimbangan antara pemenuhan standar nutrisi secara objektif dan tingkat penerimaan rasa secara subjektif. Kesalahan dalam menyusun menu seringkali berdampak pada tingginya sisa makanan akibat rasa yang kurang disukai, meskipun kandungan gizinya tinggi. Penelitian ini bertujuan untuk melakukan analisis keputusan secara ilmiah dalam mengevaluasi dan menentukan kelayakan menu makanan melalui sebuah antarmuka sistem berbasis web. Evaluasi dianalisis menggunakan algoritma logika fuzzy mamdani dengan mengolah dua variabel input utama, yaitu skor gizi komposit dan skor hedonik. Tahapan komputasi analisis meliputi proses fuzzifikasi, inferensi aturan implikasi MIN, agregasi kurva MAX dan defuzzifikasi menggunakan metode Center of Area (CoA) berbasis integral kontinu luas area. Hasil pengujian fungsionalitas menunjukkan bahwa seluruh modul komputasi beroperasi 99% valid. Analisis keputusan ini berhasil menetapkan persentase kelayakan akhir dan mengklasifikasikan menu ke dalam tiga status: Direkomendasikan, Dipertimbangkan dan Ditolak sehingga mampu memfasilitasi penentuan menu favorit secara terukur dan objektif. Kata Kunci: Analisis Keputusan, Fuzzy Mamdani, Makan Bergizi, Skor Hedonik, Angka Kecukupan Gizi.
PENGUATAN KEMAMPUAN BACA TULIS AL-QUR’AN MELALUI PROGRAM TAHSIN TERSTRUKTUR BERBASIS METODE IQRA’ DI TPA AR-ROHMAN DESA SRIMENANTI KECAMATAN TANJUNG RAJA Martoyo; Haniefah Iqtianti; Isti Koma; Julmi
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 2 (2026): April 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i2.6264

Abstract

Abstract: The ability to read and write the Qur'an is the main foundation in Islamic education that must be instilled from an early age. However, in practice, there are still many students at the Al-Qur'an Education Park (TPA) who experience difficulties in reading the Qur'an in tartil according to the rules of makharijul huruf and basic tajweed. This condition was also found in students at TPA Ar-Rohman, Srimenanti Village, Tanjung Raja District, which showed variations in the ability to read and write the Qur'an and the less than optimal implementation of structured tahsin learning. This article aims to describe the implementation and analyze the results of a community service program in the form of strengthening the ability to read and write the Qur'an through a structured tahsin program based on the Iqra' method. The method used is a Community-Based Research (CBR) mentoring approach with observation techniques, direct mentoring, and simple evaluation of the students' abilities. The program was implemented for three days with a target of 27 TPA students. The results of the activity showed an increase in the fluency and accuracy of the students' reading, especially in pronouncing the hijaiyah letters and applying basic tajweed. Furthermore, the student attendance rate reached 100%, reflecting enthusiasm and positive acceptance of the program. These findings indicate that the structured tahsin program based on the Iqra' method is effective as a model for strengthening Quranic literacy in the TPA environment and has the potential to be replicated in similar community service contexts. Keywords: Tahsin Al-Qur'an; Iqra' Method; Community Service; TPA; Quranic Literacy.   Abstrak: Kemampuan baca tulis Al-Qur’an merupakan fondasi utama dalam pendidikan Islam yang harus ditanamkan sejak usia dini. Namun demikian, pada praktiknya masih banyak santri Taman Pendidikan Al-Qur’an (TPA) yang mengalami kesulitan dalam membaca Al-Qur’an secara tartil sesuai dengan kaidah makharijul huruf dan tajwid dasar. Kondisi tersebut juga ditemukan pada santri TPA Ar-Rohman Desa Srimenanti Kecamatan Tanjung Raja, yang menunjukkan variasi kemampuan baca tulis Al-Qur’an serta belum optimalnya penerapan pembelajaran tahsin secara terstruktur. Artikel ini bertujuan untuk mendeskripsikan pelaksanaan serta menganalisis hasil program pengabdian kepada masyarakat berupa penguatan kemampuan baca tulis Al-Qur’an melalui program tahsin terstruktur berbasis metode Iqra’. Metode yang digunakan adalah pendekatan pendampingan berbasis Community-Based Research (CBR) dengan teknik observasi, pendampingan langsung, dan evaluasi sederhana terhadap kemampuan santri. Program dilaksanakan selama tiga hari dengan sasaran 27 santri TPA. Hasil kegiatan menunjukkan adanya peningkatan kelancaran dan ketepatan bacaan santri, khususnya dalam pengucapan huruf hijaiyah dan penerapan tajwid dasar. Selain itu, tingkat kehadiran santri mencapai 100%, yang mencerminkan antusiasme dan penerimaan positif terhadap program. Temuan ini menunjukkan bahwa program tahsin terstruktur berbasis metode Iqra’ efektif sebagai model penguatan literasi Al-Qur’an di lingkungan TPA dan berpotensi untuk direplikasi pada konteks pengabdian masyarakat serupa. Kata Kunci: Tahsin Al-Qur’an; Metode Iqra’; Pengabdian Kepada Masyarakat; TPA; Literasi Al-Qur’an.
PENGARUH MOTIVASI KERJA, KEPEMIMPINAN DAN LINGKUNGAN KERJA TERHADAP PRODUKTIVITAS KERJA PEGAWAI DI DINAS PERDAGANGAN DAN PERINDUSTRIAN KOTA TANJUNGBALAI Nurleli Juanda; Aris Siregar
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 2 (2026): April 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i2.6265

Abstract

Abstract: This study aims to analyze the effect of Work Motivation, Leadership, and Work Environment on Employee Work Productivity at the Department of Trade and Industry of Tanjungbalai City. The data used are primary data. This research employed a quantitative approach using multiple linear regression analysis. The results of the multiple linear regression analysis obtained the equation: Y = 0.619 + 0.379X1 + 0.366X2 + 0.328X3. The results of the simultaneous test (F test) showed that Work Motivation, Leadership, and Work Environment simultaneously have a positive and significant effect on Employee Work Productivity with an Fcount value of 16.915 > Ftable 2.75 and a significance value of 0.000 < 0.05. Partially, Work Motivation has a positive and significant effect on Employee Work Productivity with a tcount value of 2.495 > ttable 1.997 and a significance value of 0.015 < 0.05. Leadership also has a positive and significant effect with a tcount value of 4.323 > ttable 1.997 and a significance value of 0.000 < 0.05. Furthermore, Work Environment has a positive and significant effect on Employee Work Productivity with a tcount value of 2.705 > ttable 1.990 and a significance value of 0.009 < 0.05. The coefficient of determination test showed that the R Square value was 0.412, which means that 41.2% of the variation in Employee Work Productivity can be explained by Work Motivation, Leadership, and Work Environment, while the remaining 58.8% is influenced by other variables outside the research model. Keywords: Work Motivation, Leadership, Work Environment, Employee Work Productivity   Abstrak: Penelitian ini bertujuan untuk menganalisis pengaruh Motivasi Kerja, Kepemimpinan, dan Lingkungan Kerja terhadap Produktivitas Kerja Pegawai pada Dinas Perdagangan dan Perindustrian Kota Tanjungbalai. Data yang digunakan adalah data primer. Penelitian ini menggunakan pendekatan kuantitatif dengan analisis regresi linier berganda. Hasil analisis regresi linier berganda diperoleh persamaan yaitu: Y = 0,619 + 0,379X1 + 0,366X2 + 0,328X3. Hasil uji simultan (uji F) menunjukkan bahwa Motivasi Kerja, Kepemimpinan, dan Lingkungan Kerja secara simultan berpengaruh positif dan signifikan terhadap Produktivitas Kerja Pegawai dengan nilai Fhitung 16,915 > Ftabel 2,75 dan nilai signifikansi 0,000 < 0,05. Secara parsial, Motivasi Kerja berpengaruh positif dan signifikan terhadap Produktivitas Kerja Pegawai dengan nilai thitung 2,495 > ttabel 1,997 dan nilai signifikansi 0,015 < 0,05. Kepemimpinan juga berpengaruh positif dan signifikan dengan nilai thitung 4,323 > ttabel 1,997 dan nilai signifikansi 0,000 < 0,05. Selanjutnya, Lingkungan Kerja berpengaruh positif dan signifikan terhadap Produktivitas Kerja Pegawai dengan nilai thitung 2,705 > ttabel 1,990 dan nilai signifikansi 0,009 < 0,05. Nilai koefisien determinasi menunjukkan bahwa nilai R Square sebesar 0,412 yang berarti bahwa 41,2% variasi Produktivitas Kerja Pegawai dapat dijelaskan oleh Motivasi Kerja, Kepemimpinan, dan Lingkungan Kerja, sedangkan 58,8% dipengaruhi oleh variabel lain di luar model penelitian. Kata Kunci: Motivasi Kerja, Kepemimpinan, Lingkungan Kerja, Produktivitas Kerja Pegawai
PENGARUH RETURN ON ASSET (ROA), DEBT TO EQUITY RATIO (DER), DAN PRICE EARNINGS RATIO (PER) TERHADAP NILAI PERUSAHAAN SEKTOR PROPERTI DAN REAL ESTATE YANG TERDAFTAR DI BEI Syah Fitri; Hamida Sari Siregar
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 2 (2026): April 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i2.6266

Abstract

Abstract: This study was conducted to test the variables of the Influence of Return On Asset (ROA), Debt To Equity Ratio (DER) and Price Earning Ratio (PER) on the Value of Property and Real Estate sector companies listed on the IDX. With a sample of 11 companies, 3 years 2022-2024 with a total of 33 observations, quantitative research methods and using multiple linear regression. The results of the partial t-test of the Return On Asset variable have a significant effect on company value with a calculated t value> t table (7.119> 2.04227) and a significance value (0.000 <0.05), while the Debt To Equity Ratio has a significant effect on company value with a calculated t value> t table (6.705> 2.04227) and a significance value (0.000 <0.05), and the Price Earning Ratio does not have a significant effect on Company Value with a calculated t value <t table ( 1.788 < 2.04227) and significance value (0.84 > 0.05). The results of the simultaneous F test of the variables Return On Asset, Debt To Equity Ratio and Price Earning Ratio significantly influence the Company Value. producing a calculated F value> Ftable (25.013> 2.93) and a significance value (0.000 < 0.05). R2 test, the adjusted R Square value or coefficient of determination is 0.692 or 69.2%, meaning that the ability of the variables Return On Asset, Debt To Equity Ratio and Price Earning Ratio in explaining the Company Value is classified as strong, namely 69.2%, while the remaining 30.8% is explained by other factors outside the regression model. Keywords: Return On Asset (ROA), Debt To Equity Ratio (DER), Price Earning Ratio (PER), Dan Company Value   Abstrak: Penelitian ini dilakukan untuk menguji variabel Pengaruh Return On Asset (ROA), Debt To Equity Ratio (DER) dan Price Earning Ratio (PER) terhadap Nilai perusahaan sektor Properti dan Real Estate yang terdaftar di BEI. Dengan jumlah sampel 11 perusahaan, 3 tahun 2022-2024 dengan jumlah 33 pengamatan, Metode penelitian kuantitaif serta menggunakan regresi linear berganda. Hasil Uji t  secara parsial variabel Return On Asset berpengaruh signifikansi terhadap nilai perusahaan dengan nilai thitung > ttabel ( 7,119 > 2,04227 ) dan nilai signifikansi ( 0,000 < 0,05 ), sedangkan Debt To Equity Ratio berpengaruh signifikansi terhadap nilai perusahaan dengan nilai thitung > ttabel ( 6,705 > 2,04227 ) dan nilai    signifikansi ( 0,000 < 0,05 ), dan Price Earning Ratio tidak berpengaruh signifikan terhadap Nilai Perusahaan dengan nilai thitung < ttabel ( 1,788 < 2,04227 ) dan nilai signifikansi ( 0,84 > 0,05). Hasil Uji F secara simultan variabel Return On Asset, Debt To Equity Ratio dan Price Earning Ratio berpengaruh secara signifikansi terhadap Nilai Perusahaan. Menghasilkan nilai Fhitung >Ftabel (25.013 > 2,93) dan nilai signifikansi (0,000 < 0,05).Uji R2, nilai adjusted R Square atau koefisien determinasi 0,692 atau 69,2% artinya kemampuan variabel Return On Asset, Debt To Equity Ratio dan Price Earning Ratio dalam menjelaskan Nilai Perusahaan tergolong kuat yaitu 69,2% sedangkan sisanya 30,8% dijelaskan oleh faktor lain diluar model regresi. Kata Kunci: Return On Asset (ROA), Debt To Equity Ratio (DER), Price Earning Ratio (PER), Dan Nilai Perusahaan
“COMPARATIVE STUDY OF KNN AND NAIVE BAYES ALGORITHMS WITH QUESTIONNAIRE DATA FOR STUDY PROGRAM RECOMMENDATION IN THE FACULTY OF ART AND DESIGN” Surya Darma; Nita Syahputri; Nurhayati
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 2 (2026): April 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i2.6267

Abstract

Abstract: The purpose of this study is to analyze and compare the performance of the K-Nearest Neighbor (KNN) and Naïve Bayes algorithms in providing study program recommendations in the Faculty of Arts and Design. The data were obtained from 250 respondents through a questionnaire consisting of 20 indicators related to students’ interests, abilities, creativity, technology, and career preferences. The research process included data preprocessing, data transformation, dataset splitting into training and testing data, modeling using the KNN and Naïve Bayes algorithms, and model performance evaluation using accuracy metrics. The data processing was carried out using the Python programming language on the Google Colab platform. The results showed that the KNN algorithm achieved an accuracy of 94%, while the Naïve Bayes algorithm obtained an accuracy of 92%. These findings indicate that the KNN algorithm performed better in classifying study program recommendations compared to the Naïve Bayes algorithm. It is expected that this research can serve as a foundation for developing a more effective decision support system to assist prospective students in selecting study programs that match their interests and abilities. Keywords: K-Nearest Neighbor (KNN), Naïve Bayes, Machine Learning, Classification, Study Program Recommendation.   Abstrak: Tujuan dari penelitian ini adalah untuk menganalisis dan membandingkan kinerja algoritma K-Nearest Neighbor (KNN) dan Naïve Bayes dalam memberikan rekomendasi program studi di Fakultas Seni dan Desain. Data diperoleh dari 250 responden melalui kuesioner yang terdiri dari 20 indikator yang berkaitan dengan minat, kemampuan, kreativitas, teknologi, dan preferensi karier mahasiswa. Proses preprocessing data, transformasi data, pembagian dataset menjadi data pelatihan dan pengujian, pemodelan menggunakan algoritma KNN dan Naive Bayes, dan evaluasi kinerja model dengan akurasi adalah bagian dari penelitian. Proses pengolahan data dilakukan pada platform Google Colab menggunakan bahasa pemrograman Python. Hasil penelitian menunjukkan bahwa algoritma KNN memiliki akurasi sebesar 94%, sedangkan algoritma Naïve Bayes memiliki akurasi sebesar 92%. Hasil ini menunjukkan bahwa algoritma KNN lebih baik dalam mengklasifikasikan rekomendasi program studi daripada algoritma Naïve Bayes. Diharapkan penelitian ini akan menjadi dasar untuk membuat sistem pendukung keputusan yang lebih baik yang membantu calon mahasiswa memilih program studi yang sesuai dengan minat dan kemampuan mereka. Kata Kunci: KNN, Naïve Bayes, Machine Learning, Klasifikasi, Rekomendasi Program Studi.
PENGEMBANGAN INSTRUMEN EVALUASI PEMBELAJARAN MATEMATIKA: BERBASIS KOMUNIKASI DAN KONEKSI MATEMATIS Mustamin Idris; Dasa Ismaimuza; Muh. Rizal
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 7 No. 4 (2024): November 2024
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v7i4.6268

Abstract

Abstract: The aim of this research is to develop a mathematics learning outcomes instrument. The plan includes: 1) designing a math learning grid for eighth-grade junior high school, 2) a math learning outcomes test, 3) content validation by relevant math learning experts, 4) analysis of math communication instrument, and 5) analysis of math connection instrument. The research results in a communication and math connection-based junior high school math learning outcomes instrument, involving teachers and students in continuous assessment system development. The focus is on effective communication and math connection aspects. The expected indicators include the instrument's effectiveness in detecting the extent of math communication and connection in preparing and implementing the learning process. This promotes a peaceful teacher-student relationship for both written and oral communication. Keywords: Learning Evaluation, Instrument Development, Communication, Math Connection, Effective Assessment.   Abstrak: Tujuan penelitian ini adalah untuk menghasilkan rancangan pengembangan instrumen hasil belajar matematika. Rencana hasil penelitian ini dapat dideskripsikan secara ringkas 1) rancangan kisi-kisi pembelajaran matematika di tingkat SMP kelas VIII, 2) tes hasil belajar matematika, 3) Hasil validasi konten oleh pakar yang sesuai dengan kompetensi pembelajaran matematika, 4) Hasil analisis instrumen komunikasi matematika 5) Hasil analisis instrumen koneksi matematika.  Hasil penelitian ini adalah tersedianya instrumen hasil belajar matematika SMP yang berbasis komunikasi dan koneksi matematis dan penelitian ini melibatkan guru dan siswa  dalam  pengembangan instrument agar  sistem penilaian dapat berkelanjutan, dengan berfokus pada aspek komunikasi dan koneksi matematika, Indikator yang diharapkan adalah keefektifan, instrumen ini dapat digunakan untuk mendeteksi sejauh mana komunikasi matematika dan koneksi matematika dalam mempersiapkan dan melaksanakan peroses pembelajaran, sehingga hubungan guru dan siswa dalam suasana tenang untuk berkomunikasi secara tertulis maupun secara lisan. Kata Kunci: Evaluasi Pembelajaran, Pengembangan Instrumen, Komunikasi, Koneksi Matematis, Penilaian Efektif.
IT CAREER NAVIGATION: PERFORMANCE EVALUATION OF KNN AND NAÏVE BAYES IN CAREER PATH RECOMMENDATIONS FOR COMPUTER SCIENCE STUDENTS (CASE STUDY: BATTUTA UNIVERSITY) Surya Darma; Muhammad Irfan Sarif; Ahmad Jihad Alfayed; Andika Syahdewa; Katharina Tyas
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 2 (2026): April 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i2.6269

Abstract

Abstract: With the rapid development of information technology, there are many career options available in the field of informatics. However, it is often difficult for students to choose a specialization that matches their interests and abilities. The purpose of this study is to develop a career path recommendation system for informatics students and to evaluate the performance of the K-Nearest Neighbor (KNN) and Naive Bayes algorithms in classification tasks. The data used in this study were collected via a questionnaire comprising 22 assessment indicators related to students’ interests, academic understanding, and preferred work styles. A total of 300 respondent data points were utilized, with 20% allocated for testing and 80% for training. The research process included preprocessing, data transformation, modeling, and evaluation using accuracy, precision, recall, and F1-score metrics. The results show that the Naive Bayes algorithm outperforms KNN, achieving an accuracy of 97%, precision of 93%, recall of 93%, and an F1-score of 93%. Therefore, Naive Bayes is considered more optimal in terms of classification performance. It is expected that the developed system can assist students in determining their career paths in a more data-driven and objective manner. Keywords: Machine Learning, KNN, Naïve Bayes, Career Recommendation, Classification   Abstrak: Dengan berkembangnya teknologi informasi yang cepat, ada banyak pilihan karir di bidang informatika. Namun, sulit bagi mahasiswa untuk memilih spesialisasi yang sesuai dengan minat dan kemampuan mereka. Tujuan dari penelitian ini adalah untuk membuat sistem rekomendasi jalur karier untuk mahasiswa informatika dan juga untuk mengevaluasi bagaimana algoritma K-Nearest Neighbor (KNN) dan Naive Bayes bekerja dalam klasifikasi. Data yang digunakan diperoleh melalui kuesioner yang terdiri dari 22 indikator penilaian yang berkaitan dengan minat mahasiswa, pemahaman akademik, dan gaya kerja yang mereka sukai. Sebanyak 300 data dari responden digunakan, dengan 20% data dialokasikan untuk pengujian dan 80% untuk pelatihan. Proses penelitian termasuk tahapan preprocessing, transformasi data, pemodelan, dan evaluasi menggunakan metrik akurasi, presisi, recall, dan skor F1. Hasil penelitian menunjukkan bahwa algoritma Naïve Bayes lebih baik dibandingkan KNN dengan nilai akurasi 97%, presisi 93%, recall 93%, dan skor F1. Akibatnya, Naïve Bayes lebih optimal dalam member. Diharapkan sistem yang dibuat dapat membantu mahasiswa dalam menentukan karir mereka secara lebih berbasis data dan objektif. Kata Kunci: Machine Learning, KNN, Naïve Bayes, Rekomendasi Karier, Klasifikasi
KLASIFIKASI SENTIMEN BERBASIS MACHINE LEARNING TERHADAP PENONTON “WONDERLAND INDONESIA” PADA PLATFORM YOUTUBE Surya Darma; Linda Wahyuni; Nauval Alfarizi
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 2 (2026): April 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i2.6270

Abstract

Abstract: The purpose of this study is to analyze and classify audience sentiment toward the Wonderland Indonesia video on the YouTube platform using the Naive Bayes and Support Vector Machine (SVM) algorithms. The dataset consists of 10,000 comments collected through a scraping process. The data were then further processed using the Sastrawi library, including data cleaning, case folding, tokenization, stopword removal, and stemming. The TF-IDF method with a dimensional size of (9048, 5000) was used for feature representation. The data distribution indicates a class imbalance, where neutral (4,550) and positive (4,297) sentiments dominate compared to negative (201). The experimental results show that SVM outperforms Naive Bayes, achieving an accuracy of 93% compared to 86%. However, both models encounter difficulties in identifying the negative class. The use of imbalanced data handling techniques shows that SVM with class weighting improves the detection of minority classes without reducing overall accuracy. In contrast, Naive Bayes with SMOTE improves recall but reduces overall performance. The findings indicate that SVM with class weighting is the best-performing model for sentiment classification. Additionally, the results show that the majority of comments are positive, suggesting that audiences respond favorably to the Wonderland Indonesia content. Keywords: Sentiment Analysis, Naive Bayes, Support Vector Machine, YouTube, TF-IDF Abstrak: Tujuan penelitian ini adalah untuk menganalisis dan mengklasifikasikan sentimen penonton terhadap video Wonderland Indonesia yang ditampilkan di platform YouTube menggunakan algoritma Naive Bayes dan Support Vector Machine (SVM). Data yang digunakan terdiri dari 10.000 komentar yang dikumpulkan melalui proses scraping. Kemudian, data diproses lebih lanjut dengan menggunakan pustaka Sastrawi untuk membersihkan, menggabungkan case, tokenize, menghilangkan stopword, dan stemming. Metode TF-IDF dengan dimensi (9048, 5000) digunakan untuk mempresentasikan fitur. Distribusi data menunjukkan ketidakseimbangan kelas; sentimen netral (4.550) dan positif (4.297) mendominasi dibandingkan dengan negatif (201). Hasil pengujian menunjukkan bahwa SVM unggul dengan akurasi 93% dibandingkan Naive Bayes sebesar 86%. Namun, kedua model kesulitan menemukan kelas negatif. SVM dengan berat kelas dapat meningkatkan deteksi kelas minoritas tanpa mengurangi akurasi, seperti yang ditunjukkan oleh penggunaan teknik penanganan data tidak seimbang. Di sisi lain, Naive Bayes dengan SMOTE meningkatkan recall tetapi mengurangi performa keseluruhan. Hasil penelitian menunjukkan bahwa model terbaik untuk klasifikasi sentimen adalah SVM dengan berat kelas. Hasil penelitian juga menunjukkan bahwa sebagian besar komentar bersentimen positif, yang menunjukkan bahwa audiens menyukai konten Wonderland Indonesia. Kata Kunci: Analisis Sentimen, Naive Bayes, SVM, YouTube, TF-IDF
Tantangan dan Peluang Teknologi Digital Bagi Eksistensi Budaya Kalwedo dan Implikasinya dalam Dunia Pendidikan Febrilien Matresya Matulatuwa; Amos Lekiwona
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 9 No. 4 (2026): August 2026 (1)
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i4.6272

Abstract

In the era of technology that continues to develop, the people of Southwest Maluku who live in areas classified as 3T (Underdeveloped, Frontier and Outermost) face significant challenges related to the cultural influence offered through online media which has become a necessity in people's lives. The purpose of this research is to discuss the importance of understanding the challenges and opportunities of digital technology for the cultural existence of the Moa people. The method used in this study is qualitative because the data generated is descriptive in nature. The results of the study relate to the challenges and opportunities of digital technology for the existence of kalwedo culture in the context of the life of the Moa community which has experienced a significant decline, caused by foreign culture offered through digital technology which is more popular with almost all young people who are the successors for sustainable regional development. People no longer respect each other, no longer build direct communication and even people who are close feel far and people who are far feel close. This kind of living condition was never imagined by the ancestors who tried to create kalwedo culture as the identity of the Moa community in particular but also the people of Southwest Maluku in general. The world of education is also unable to design a curriculum that can answer the challenge of the fading value of kalwedo culture as something that needs to be preserved through the learning process

Page 327 of 365 | Total Record : 3645

 325   326   327   328   329 

Filter by Year

2018 2026


Reset
Filter By Issues
All Issue Vol. 9 No. 4 (2026): August 2026 (1) Vol. 9 No. 3 (2026): June 2026 Vol. 9 No. 2 (2026): April 2026 Vol. 9 No. 1 (2026): February 2026 Vol 9, No 1 (2026): February 2026 Vol. 8 No. 4 (2025): November 2025 Vol 8, No 4 (2025): November 2025 Vol 8, No 3 (2025): August 2025 Vol. 8 No. 3 (2025): August 2025 Vol. 8 No. 2 (2025): May 2025 Vol 8, No 2 (2025): May 2025 Vol. 8 No. 1 (2025): February 2025 Vol 8, No 1 (2025): February 2025 Vol 7, No 4 (2024): November 2024 Vol. 7 No. 4 (2024): November 2024 Vol 7, No 3 (2024): August 2024 Vol. 7 No. 3 (2024): August 2024 Vol 7, No 2 (2024): May 2024 Vol. 7 No. 2 (2024): May 2024 Vol 7, No 1 (2024): February 2024 Vol. 7 No. 1 (2024): February 2024 Vol 6, No 3 (2023): October 2023 Vol. 6 No. 3 (2023): October 2023 Vol 6, No 2 (2023): June 2023 Vol 6, No 1 (2023): February 2023 Vol. 5 No. 3 (2022): October 2022 Vol 5, No 3 (2022): October 2022 Vol 5, No 2 (2022): June 2022 Vol 5, No 1 (2022): February 2022 Vol 4, No 3 (2021): October 2021 Vol 4, No 2 (2021): June 2021 Vol 4, No 1 (2021): February 2021 Vol 3, No 2 (2020): August 2020 Vol 3, No 1 (2020): February 2020 Vol 2, No 2 (2019): August 2019 Vol 2, No 1 (2019): February 2019 Vol 1, No 2 (2018): August 2018 Vol 1, No 2 (2018): August 2018 Vol 1, No 1 (2018): February 2018 Vol 1, No 1 (2018): February 2018 More Issue