Danilo Dorado

Prof. Danilo Dorado

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 May 2024 in Information
Verified
Journal Article
Open access
PDF file
QWQinyong Wang
Qinyong Wang
EJEnman Jin
Enman Jin
HZHuizhong Zhang
Huizhong Zhang
YCYumeng Chen
Yumeng Chen
YYYinggao Yue
Yinggao Yue
MXMinghai Xu
Minghai Xu
Personalized recommender systems play a crucial role in assisting users in discovering items of interest from vast amounts of information across various domains. However, developing accurate personalized recommender systems remains challenging due to the need to balance model architectures, input feature combinations, and fusion of heterogeneous data sources. This study investigates the impacts of these factors on recommendation performance using the MovieLens and Book Recommendation datasets. Six models, including single-task neural networks, multi-task learning, and baselines, were evaluated with various input feature combinations using Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). The multi-task learning approach achieved significantly lower RMSE and MAE by effectively leveraging heterogeneous data sources for personalized recommendations through a shared neural network architecture. Furthermore, incorporating user data and content data progressively enhanced performance compared to using only item identifiers. The findings highlight the importance of advanced model architectures and fusing heterogeneous data sources for high-quality recommendations, providing valuable insights for designing effective recommender systems across diverse domains.

26 July 2024
Verified
Conference Paper
Open access
DDDanilo Dorado
Danilo Dorado
JAJoey Aviles
Joey Aviles
Road accidents contribute significantly to annual fatalities. Various agencies engaged in road safety management are consistently striving for quality improvement to predict and mitigate accidents through the aid of information technology. As machine learning emerged, its data analysis, visualization, and prediction capabilities have proven to help policymakers. The main objective of this study is to develop a machine-learning model for predicting road accidents in Urdaneta City, Philippines. The dataset used consists of three years (2021–2023) of road accident records from the Emergency Medical Services (EMS). Various preprocessing methods are employed, and a grid search technique is implemented to identify optimal features for the model. Random Forest, XGBoost, AdaBoost, Decision Tree, and LGBM algorithms are developed to predict and visualize road accident occurrences. Upon development, it was found that XGBoost performs well among other models, acquiring a total Rsquared of 0.9996 in training and 0.9898 during testing. To validate the findings of the study, the model's interpretability was assessed using a William Plot. The SHAP and LIME methodologies offer valuable insights into how various variables affect the model's forecast of the monthly number of road accidents. The Permutation Feature Importance approach provides insight into the specific impact of each feature on the model's predictions, emphasizing the significant influence of time on accident frequency, particularly during peak hours or late-night accidents. The Box Plot offers valuable insights, with the median (Q2) indicated by a line showing a value of around 10. Leveraging this model can significantly reduce the likelihood of accidents in real-time, resulting in a safer and more intelligent transportation ecosystem.