Gunadarma University

Accurate housing price estimation is essential for supporting real estate decision making and urban economic planning.This study proposes an integrated framework that combines ensemble machine learning models with a Natural Language Processing (NLP) based conversational interface for housing price prediction and price-zone classification in the JABODETABEK region.A dataset of 3,553 property listings was preprocessed through data cleaning, missing value handling, outlier detection using the Interquartile Range (IQR) method, logarithmic transformation, and feature engineering.Comparative experiments were conducted using Linear Regression, Random Forest, Gradient Boosting, and XGBoost for regression tasks, and Random Forest, Decision Tree, K-Nearest Neighbors, and Gradient Boosting for classification tasks.XGBoost achieved the best regression performance with approximately 96% predictive accuracy, while Random Forest demonstrated superior classification performance with an accuracy of 87.46%.The NLP intent classification module, developed using a Bag-of-Words representation and Multinomial Nave Bayes, achieved 94.82% training accuracy and 90.20% testing accuracy.All components were integrated into a Command Line Interface (CLI)-based chatbot capable of interpreting user queries and generating automated price estimations and price-zone classifications.The results demonstrate that the proposed unified framework provides robust predictive performance while enhancing user accessibility through conversational interaction.

Background: The increasing environmental challenges and sustainability pressures have significantly transformed the way firms formulate their competitive strategies. In response to these global changes, green entrepreneurial orientation and environmental orientation have emerged as critical organizational capabilities that enable businesses to balance economic performance with societal and environmental responsibilities. Despite the growing body of research in this area, existing studies remain fragmented and lack an integrated understanding of the intellectual structure and development trajectory of this research field. Objective: This study aims to map the global research landscape of green entrepreneurial orientation and environmental orientation through a comprehensive bibliometric analysis grounded in the Natural Resource-Based View (NRBV). Methods: Using publications retrieved from the Semantic Scholar database from 2010 to 2025, bibliometric techniques were employed to analyze publication trends, influential authors and journals, collaborative networks, and thematic evolution using VOSviewer software. Results: The findings reveal rapid growth of research interest, with dominant themes focusing on green innovation, sustainable competitive advantage, and business performance. Network, overlay, and density visualizations indicate strong intellectual linkages between entrepreneurial orientation, environmental strategy, and firm-level outcomes, reflecting a recent shift towards innovation-driven sustainability. Conclusion: This study provides an integrated overview of sustainability-oriented strategic orientations and their role in supporting business resilience, offering implications for managers, policymakers, and researchers aligned with sustainable development goals.

This study aims to find out and analyze how effective visual content is in promotion strategies in the Industrial Revolution Era 4.0 on Instagram @Beanspot_id as well as knowing the respondents' responses about the level of awareness, comprehension, intention, interest, action on visual content in promotional strategies on Instagram @beanspot_id. This study used a quantitative method with a case study approach and a positivism paradigm. Data was obtained through the method of distributing online questionnaires, namely the Google Form which was carried out to 100 respondents with the criteria, namely followers of the Instagram account @beanspot_id. The results of this study reveal that through calculations using the Customer Response Index (CRI) method, which measures the effectiveness of visual content in promotional strategies in the industrial revolution era 4.0 on Instagram @beanspot_id, is less effective. The recommendation for Beanspot, especially the marketing team, to be able to explore more creative ideas and optimize more in terms of visualization so that they can attract new followers then increase the brand awareness, as well as the sales, especially on social media. For the development of further research, researchers provide suggestions using the CRI method on the Instagram Beanspot, which was declared to be ineffective, this could occur because of several factors, one of which is because there is a lot of competition via Instagram which is quite tight for similar coffee to go brands. Thus, the researcher’s suggestion for Beanspot, especially the marketing team, is to be able to explore more creative ideas and optimize more in terms of visualization such as making quizzes with prizes that can attract new followers to increase brand awareness as well as being able to develop sales, especially on social media.

Solar energy is a renewable energy source that can serve as an alternative for electricity generation. Indonesia, as a tropical country, possesses significant solar energy potential and is also an archipelagic nation rich in marine resources. One of these resources is seaweed, a multifunctional marine product that requires drying. A seaweed drying machine utilizes solar panel energy as its power source, enabling an efficient and effective drying process under various weather conditions. The research method employed involved a series of calculations followed by data analysis derived from the results, along with literature review, discussion, and field observation. The analysis was conducted using a 20 Wp solar panel with tilt angle variations of 0°, 15°, and 30°, oriented southward. The experiment was carried out over a 9-hour period each day, from 07:00 to 15:00 local time (WIB), spanning from January 15 to March 8, 2024. The results demonstrate that the highest power efficiency was achieved with a photovoltaic module tilt angle of 15°, reaching 99.47%, with an average output power (Pout) of 4.53 Watts.

This study aims to explore the role of public information transparency in building community trust toward organizational sustainability initiatives. With the increasing attention to sustainability practices, it is crucial for organizations to alter their information dissemination strategies. This qualitative research employed in-depth interviews with 10 participants, consisting of organizational representatives and community members. The analysis results indicate that a high level of information transparency positively impacts community trust. The findings also suggest that transparency in communication, both through sustainability reports and two-way dialogue, plays a significant role in reducing scepticism and enhancing community engagement. The study concludes that organizations prioritizing information transparency can establish stronger relationships with the public and improve the effectiveness of their sustainability initiatives. Therefore, it is recommended that organizations implement more transparent communication strategies to maximize public trust and support for their sustainability efforts.

Social media has become a major space for sharing content and opinions. In online settings, individuals may show behaviors they would usually inhibit in face-to-face interactions, known as the online disinhibition effect. Understanding this tendency among early adults is important because this group is highly active on social media and may be exposed to both constructive and harmful digital interactions. Objective: This study aims to describe the characteristics and level of the online disinhibition effect among early adult social media users. Methods: A quantitative descriptive design was used with 111 early adults (20–40 years; males and females) recruited through non-probability incidental sampling. Data were collected via an online questionnaire (Google Forms) using an adapted Online Disinhibition Scale based on Suler’s dimensions, complemented by demographic items (gender, age, daily duration of use) and two open-ended questions. The instrument was evaluated for validity, item discrimination, and reliability. Data were analyzed using descriptive statistics and frequency-based categorization for open-ended responses. Results: The online disinhibition effect was categorized as high (empirical mean = 85.50; hypothetical mean = 73.50). Male participants showed higher disinhibition than female participants. Higher disinhibition was also observed among participants who used social media longer each day, especially those reporting more than six hours of use. Open-ended responses indicated that anonymity and the absence of face-to-face interaction were common reasons for feeling freer online, and hostile commenting was frequently reported as an activity more comfortably performed on social media than in real life. Implications: Findings support the need for digital well-being efforts targeting early adults, including self-regulation strategies (e.g., screen-time limits, reflection-before-post routines) and digital civility education, alongside platform-level measures that strengthen accountability cues. Originality: This study provides a descriptive baseline of online disinhibition among Indonesian early adults using a psychometrically screened measure and triangulates findings with open-ended evidence linked to key theoretical dimensions.

Kemudahan akses pinjaman online (pinjol) mendorong peningkatan penggunaan layanan kredit digital, namun sering diikuti perilaku berutang yang kurang terencana akibat minimnya pemahaman biaya, tenor, dan risiko, termasuk risiko layanan ilegal dan penyalahgunaan data pribadi. Penelitian ini bertujuan menganalisis perubahan perilaku berutang peserta setelah edukasi “Baca–Hitung–Cek sebelum Klik Setuju” sebagai strategi penggunaan pinjol legal yang aman dan bertanggung jawab. Metodologi menggunakan pendekatan campuran (mixed methods) dengan desain pretest–posttest dan pendalaman kualitatif. Responden merupakan peserta kegiatan edukasi literasi keuangan digital pada komunitas/instansi mitra. Data kuantitatif dikumpulkan melalui kuesioner yang mengukur pemahaman struktur biaya (bunga, biaya layanan, denda), kemampuan perencanaan pembayaran (cashflow sederhana), perilaku verifikasi legalitas, serta indikator kehati-hatian digital (izin aplikasi, OTP, phishing). Data kualitatif diperoleh melalui wawancara singkat/FGD untuk menangkap alasan perubahan keputusan berutang dan hambatan penerapan strategi. Analisis kuantitatif dilakukan dengan statistik deskriptif dan uji beda berpasangan, sedangkan data kualitatif dianalisis tematik. Temuan menunjukkan peningkatan signifikan pada pemahaman komponen biaya dan kemampuan menghitung total kewajiban, peningkatan kepatuhan verifikasi legalitas sebelum meminjam, serta penurunan kecenderungan mengambil pinjaman tanpa perencanaan. Peserta juga melaporkan perubahan strategi penggunaan pinjol menjadi lebih selektif, seperti membatasi nominal, memilih tenor realistis, dan mengutamakan kebutuhan produktif. Implikasi penelitian menegaskan bahwa strategi edukasi sederhana dan operasional “Baca–Hitung–Cek” efektif mendorong perilaku pinjol yang lebih aman, dan dapat dijadikan model intervensi literasi keuangan digital berbasis komunitas/sekolah. Orisinalitas penelitian terletak pada pengukuran perubahan perilaku berutang yang dikaitkan langsung dengan langkah praktis pengambilan keputusan pinjol (membaca ringkasan, menghitung total biaya, dan mengecek legalitas) serta integrasinya dengan praktik keamanan data pribadi.

The financial performance of an entity serves as a measure used to assess a company's ability to generate revenue or profits reflected in a given period. One of the digital banks in Indonesia is Allo Bank. The purpose of this research is to provide an analysis of how the assessment of the health level of PT Allo Bank Indonesia Tbk for the 2020-2024 period uses eight variables, namely NPL, LDR, GCG, ROA, ROE, NIM, BOPO, and CAR. The analysis technique applied in this study is to analyze financial statements through the RGEC approach. The results of this study show that overall, during the period 2020 to 2024, Allo Bank's health condition received a Very Healthy rating in 2020 and 2021, with composite values of 87.50% and 90%, respectively. And obtained the Healthy rating for the 2022-2024 period with a composite value of 80%, 82.50%, and 82.50%, respectively.

App-store reviews are a rich source of user feedback for improving mobile service quality, yet they are large in volume, unstructured, and can change rapidly after version updates, making manual triage difficult to perform consistently. This condition requires an approach that can transform review text into prioritized, decision-ready information through a monitoring-oriented information system. Objective: This study aims to design and evaluate an app-review analytics information system that integrates sentiment classification and topic modeling to capture user perception, identify dominant issues, and support faster and more measurable prioritization of app improvements. Methodology: The study employs a quantitative design using an applied case study and a design-and-evaluate approach. Secondary data were collected from 5,000 Google Play reviews over 12 months (text, rating, timestamp), and primary data were obtained through dashboard usability testing with 20 participants using the System Usability Scale (SUS). The analysis includes text preprocessing, TF–IDF-based sentiment classification with multiple model comparisons, LDA topic modeling to extract recurring issues, and integration of outputs into a dashboard for trend monitoring and reporting. Findings: The best-performing model (SVM) achieved Accuracy = 0.86 and Macro-F1 = 0.84 for three-class sentiment classification (positive–neutral–negative). Topic modeling produced 10 dominant topics, with negative sentiment most strongly concentrated in core service issues such as login/OTP failures (72% negative), post-update crashes (69%), payment/transaction errors (65%), and server downtime (63%). The dashboard achieved a usability score of SUS = 82.3, indicating strong operational readiness for monitoring and triage. Implications: The results provide a practical basis for app managers to conduct periodic monitoring, convert dominant complaints into prioritized maintenance backlogs, and assess the impact of updates through sentiment and topic trends. The approach can be adopted as a routine decision-support mechanism for product owners, helpdesk teams, and developers. Originality/Value: This study’s novelty lies in its end-to-end integration of NLP analytics (sentiment and topics) with an information system artifact (dashboard), complemented by standardized usability validation. Unlike studies that stop at model outputs, this work demonstrates how analytics can be operationalized into a continuously usable monitoring system for app-maintenance decision-making.

Penelitian ini bertujuan untuk menganalisis determinan keputusan bertransaksi menggunakan QRIS, yaitu literasikeuangan,persepsikemudahan, persepsi kepercayaan, persepsi keamanan terhadap keputusan bertransaksi menggunakan QRIS dan untukmengetahuivariable yang paalingdominan mempengaruhi keputusan bertransaksi menggunakan QRIS. Pengumpulandatadalampenelitianinimenggunakan kuesioner yang disebar melalui g-form kepada 100 responden. Teknik pengambilan sampel yang digunakan adalah nonprobabilitysamplingdenganpurposivesampling.Teknik analisis menggunakan, uji validitas, uji reliabilitas, uji asumsi klasik dan uji regresi linier berganda. Hasil penelitian menunjukkan bahwa literasi keuangan, persepsi kemudahan, dan persepsi keamanan berpengaruh terhadap keputusan bertransaksi menggunakan QRIS. Sedanglan persepsi kepercayaan tidak berpengaruh terhadap keputusan bertransaksi menggunakan QRIS. Persepsi keamanan adalah variable yang paling dominan mempengaruhi keputusan bertransaksi menggunakan QRIS.