Jurnal Indonesia Sosial Teknologi

Pengaruh Pelatihan dan Pengembangan Karir terhadap Kinerja Karyawan Ritel Lokal: Studi Kasus pada Karyawan X-Sha Kabupaten Tasikmalaya

2026-10-17T11:20:58+00:00

Latar Belakang: Pengelolaan modal manusia secara optimal melalui pelatihan dan pengembangan karir sangat penting bagi perusahaan ritel untuk menjaga kinerja, namun kendala operasional sering menghambat jadwal pelatihan dan objektivitas promosi. Tujuan: Penelitian ini bertujuan untuk menganalisis pengaruh pelatihan dan pengembangan karir secara parsial dan simultan terhadap kinerja karyawan X-Sha, sebuah perusahaan ritel lokal di Kabupaten Tasikmalaya. Metode: Penelitian ini menggunakan pendekatan kuantitatif dengan desain survei potong lintang (cross-sectional). Sampel berjumlah 112 dari 155 karyawan di 15 gerai yang ditentukan dengan rumus Yamane dan dipilih melalui teknik proportionate stratified random sampling. Data dikumpulkan menggunakan kuesioner skala Likert lima poin, ditransformasi dari skala ordinal ke interval menggunakan Method of Successive Interval (MSI), kemudian dianalisis dengan regresi linier berganda pada IBM SPSS 27 setelah melalui uji instrumen dan uji asumsi klasik. Hasil: Hasil analisis deskriptif menunjukkan bahwa pelatihan dan pengembangan karir berada pada kategori sangat tinggi, sedangkan kinerja karyawan berada pada kategori tinggi. Secara parsial, pelatihan (t = 3,613; p = 0,044) dan pengembangan karir (t = 5,969; p = 0,004) berpengaruh positif dan signifikan terhadap kinerja karyawan, dengan pengembangan karir sebagai prediktor paling dominan (β = 0,559). Secara simultan, kedua variabel berpengaruh signifikan (F = 21,627; p = 0,000) dan menjelaskan 47,6% variasi kinerja (R² = 0,476). Kesimpulan: Pelatihan yang terencana dan jalur karir yang terstruktur berhubungan positif dengan kinerja karyawan pada bisnis ritel lokal. Implikasinya, manajemen perlu mengintegrasikan hasil evaluasi pelatihan ke dalam standar operasional promosi karir yang objektif.

Analisis Kendala Literasi Digital dalam Pemasaran UMKM dan Sektor Pertanian di Desa Linggajati

2026-10-17T10:58:17+00:00

The digitalization of marketing contributes to enhancing the competitiveness of Micro, Small, and Medium Enterprises (MSMEs) and the agricultural sector in rural areas; however, limited digital literacy remains a primary obstacle for local business owners in leveraging technology effectively. This study addresses the challenges related to digital literacy in marketing among MSMEs and the agricultural sector in Linggajati Village. The research focuses on the use of digital technology for marketing activities, specifically within community-based food MSMEs and the agricultural sector. Employing a qualitative descriptive method and a case study approach, data were collected through observations, interviews, and questionnaires involving eight MSME owners and two farmers. The findings indicate that the main challenges faced by these MSME owners and farmers include low digital literacy, limited use of marketing technology, and a continued reliance on traditional promotional methods. Most MSME owners use only WhatsApp for marketing; they do not utilize online marketplaces, have not adopted digital payment systems (such as QRIS), and are not actively managing social media. In the agricultural sector, the marketing of harvests remains conventional and is not yet integrated with digital platforms. Factors such as age, educational background, limited capital, and a lack of digital training are the primary causes of this situation. These findings suggest that the low level of digital literacy among MSME owners and farmers necessitates sustained guidance and digital training, tailored to the specific characteristics of the community in Linggajati Village. As a recommendation, future research should involve a broader range of respondents and examine the effectiveness of digital training programs in enhancing technology-based marketing within rural areas.

A Comparative Study of Machine Learning Algorithms for Indonesian Hoax Detection on Social Media

2026-09-11T01:50:19+00:00

Social media platforms have accelerated the spread of misinformation (hoax) in Indonesia, undermining public trust and informed decision-making. This study aims to compare the performance of three text classification algorithms in detecting Indonesian-language hoax content on social media and to identify linguistic patterns distinguishing hoax from valid content. A publicly available corpus of 3,754 Indonesian tweets labeled as hoax or valid was used. Text was preprocessed through case folding, URL removal, and non-alphabetic character removal, then represented using TF-IDF with unigram and bigram features. Three classifiers, Multinomial Naive Bayes, Logistic Regression, and Linear Support Vector Machine, were trained and evaluated on an 80:20 stratified train-test split using accuracy, precision, recall, F1-score, and Area Under the ROC Curve (AUC). Results: Linear SVM achieved the highest accuracy (84.29%) and F1-score (0.9106), Logistic Regression achieved the highest precision (0.9118) and AUC (0.8156), while Naive Bayes achieved the highest recall (0.9967). Feature analysis revealed that debunking-related terms such as hoax, hoaks, mitos, and faktanya were strongly associated with the valid class, since such terms frequently appear in fact-checking tweets rather than in genuinely neutral news content. Conclusion: Model selection for hoax detection should be guided by the operational goal, favoring high-recall models such as Naive Bayes when the priority is minimizing missed hoaxes, and higher-precision models such as Logistic Regression when minimizing false accusations is prioritized; additionally, the reliance on debunking-style vocabulary as a valid-class signal highlights an important limitation of purely lexical models that future research should address.

Explainable Multi-Source Machine Learning for PM₂.₅ Mapping Across Indonesian Cities: Integrating Satellite, Meteorological, Land-Surface and Fire Information

2026-10-17T11:07:27+00:00

Background: Fine particulate matter (PM₂.₅) is a critical indicator of atmospheric environmental quality, yet its spatial assessment in Indonesia remains constrained by the limited and uneven distribution of ground-monitoring stations. Objectives: This study develops an explainable multi-source machine-learning framework for estimating daily PM₂.₅ concentrations across Indonesian cities by integrating ground observations with satellite-derived aerosol optical depth, meteorological reanalysis, land-surface characteristics, and satellite-detected fire activity. Method: Ground-level PM₂.₅ observations are harmonized with MODIS MAIAC aerosol data, ERA5-Land meteorological variables, VIIRS fire information, and selected geographic covariates. Linear baseline models, Random Forest, and gradient-boosted trees are evaluated using temporal-block validation and leave-one-city-out cross-validation to distinguish temporal predictive performance from geographic transferability. SHapley Additive exPlanations (SHAP) are used to identify predictor contributions, while spatial residual diagnostics are employed to detect systematic geographic prediction errors. Results: In the present simulated analytical scenario, the dataset comprised 3,854 station-days from six Indonesian cities during 2023–2024. The best-performing XGBoost model achieved an out-of-sample R² of 0.79, RMSE of 7.8 µg m⁻³, and MAE of 5.6 µg m⁻³ under temporal validation. Under leave-one-city-out validation, performance decreased to R² = 0.64, RMSE = 10.5 µg m⁻³, and MAE = 7.7 µg m⁻³. Satellite AOD, planetary boundary-layer height, relative humidity, wind speed, and lagged fire radiative power were the dominant predictors. Conclusion: The proposed framework provides a transparent approach for extending PM₂.₅ assessment beyond existing monitoring locations while explicitly evaluating geographic generalizability and model interpretability.

Implementation of Contrast Limited Adaptive Histogram Equalisation in CNN-Based Pneumonia Detection Using Chest X-ray Images: A Systematic Literature Review

2026-10-17T11:01:06+00:00

Pneumonia remains a major global health burden, and chest X-ray (CXR) interpretation is constrained by image quality and observer variability. Contrast Limited Adaptive Histogram Equalisation (CLAHE) is widely used to enhance local contrast before convolutional neural network (CNN) classification, yet its contribution has not been synthesised systematically. This study aims to review how CLAHE is implemented in CNN-based pneumonia detection, which architectures accompany it, how it relates to diagnostic performance, and which methodological gaps remain. A systematic literature review following the PRISMA 2020 guidelines was conducted on the Scopus database for 2021–2026; of 82 records identified, 21 journal articles were included and analysed through thematic synthesis. Four themes emerged: CLAHE implementation strategies, CNN architectures and learning strategies, performance of CLAHE-based pipelines, and methodological robustness. CLAHE was rarely applied alone; it was combined with normalisation, denoising, and augmentation and was increasingly optimised or modified. Transfer learning models dominated, and the highest reported accuracies ranged from 92% to 99.78%. However, heterogeneous datasets, CLAHE parameters, validation protocols, and metrics, together with scarce ablation and external validation, prevent attributing these gains to CLAHE alone. Standardised preprocessing, ablation-based reporting, multicentre validation, and explainable AI are needed to make CLAHE-based CNN systems clinically reliable.