The use of information technology is very important to support speed and ease in serving the community. However, at the village level there is still a shortage of specialized staff in the field of providing information technology-based applications. So, it is necessary to do community service in the form of technical guidance to take advantage of the use of applications such as Google workspace. Google workspace makes it very easy to manage data and information management. The data collection process is carried out using the Google form, data processing uses Google sheets, and administrative work can use the collaboration feature on Google Docs. This community service program is intended to explore Google workspace facilities, in creating data visualization dashboards using Google Data Studio. Villagers can monitor the movement of data in real-time against input from the Google form in the form of interactive graphics so that it can be used as a basis for making decisions more quickly.

This community service program aims to improve the quality of management and promotion of Jumpinang River Tubing tourist destinations in Pasuruan through the use of digital technology. Implementation methods include problem identification, website creation, and guiding and safety training for local communities. This process was carried out for 3 months, with core activities in the form of Focus Group Discussion (FGD), website development, and technical training. The results of the program show that the community's capacity is increased in the management of tourist destinations, as well as the ease of promotion through digital platforms. The implication of this program is the increase in the number of tourist visits and the strengthening of the sustainability of tourism management by the community.

An End-to-End Machine Learning Pipeline for Online Purchase Intention Prediction Using Random Forest and MLOps Practices

Background: Social-media-based risk screening is promising, yet deployment quality is constrained by noisy language, class imbalance, low explainability, and uncertainty handling. Objective: This work develops an application-specific BiLSTM-centered proxy risk screening framework that combines chaos-regularized learning, SMOTE balancing, SHAP explanation, and TF-IDF similarity fallback. Methods: Experiments used an annotated Twitter corpus (>31,000 English posts) with neutral, hate, and offensive labels, where hate/offensive content was treated only as a proxy risk indicator rather than as evidence of depression, self-harm, or clinical psychological distress. The workflow covered text normalization, tokenization, fixed-length sequence modeling, imbalance correction, and thresholded inference. The model stack used GloVe embeddings with recurrent layers and chaos-based dropout, trained with Adam plus adaptive stopping/scheduling. Operational safety was supported through lexicon checks and a balanced TF-IDF reference set for low-confidence cases. Results: Validation performance reached an F1-score of 0.86, accuracy 0.79, and AUC 0.93 under the reported pipeline. Component-level evidence further indicated that SHAP, thresholding, TF-IDF retrieval, and lexicon checks add explanation, uncertainty handling, reference context, and conservative escalation beyond the underlying classifier, although their marginal effects were not isolated through rerun ablation. Conclusion: The proposed system offers a practical tradeoff between predictive quality, interpretability, and safety-aware operation for proxy risk screening on noisy social media. The contribution is positioned as an integrated application pipeline rather than a new NLP algorithm, and future work should validate the framework with clinically grounded labels, leakage-safe resampling, ablation studies, and transformer baselines.