Research

As the national university, we champion and support innovative research that addresses the country’s most pressing challenges.

24 Jan 2024

UP Manila

People who are new patients or seeking clearance or referral for endocrine-related symptoms are more likely to opt for teleconsultation

Telemedicine employs the use of technology to increase access to health care. This is especially relevant in developing countries where...

Read More

23 Jan 2024

UP Diliman

While Visayan women were depicted as “most civilized” in colonial photography, the photos still upheld a gender ideology promoting male superiority

Significant works have been published on American colonial photography in the Philippines, which primarily focus on the depiction of Filipinos...

Read More

22 Jan 2024

UP Manila

To improve how we diagnose, prevent and treat allergies, it is crucial to better understand how different factors interact with each other

Allergies are overreactions of the body to harmless allergens and are one of the most chronic conditions worldwide. Atopy or...

Read More

18 Jan 2024

UP Los Baños

A majority of world’s largest flowers teeter on the edge of extinction

The genus Rafflesia, which includes the world’s largest flowers, has aroused curiosity among scientists for centuries and features prominently in local...

Read More

17 Jan 2024

UP Mindanao

Utilizing chitosan as an edible coating presents a promising solution for prolonging the shelf life of fresh produce

The development of diseases and decay in vegetables after harvest limits their shelf life and saleability. To control this, proper...

Read More

16 Jan 2024

UP Diliman

An easy-to-build computer-based spellchecker designed to help foreign language learners in writing provides error-specific feedback with explanations and examples

In second language (L2) acquisition, writing is no longer perceived as a mere consequence of learning, but is now treated...

Read More

15 Jan 2024

UP Manila, UP Open University

A novel method has the capability to precisely identify the author of unknown texts, particularly news articles

This study tackles the challenge of figuring out who wrote anonymous text, like online articles, in an age when there...

Read More

11 Jan 2024

UP Manila

The main obstacles to a “good death” during the Covid-19 pandemic were strict limitations on comforting and communicating with patients

Dying is not a medical experience but also a social and psychological one. We would like to ease the suffering...

Read More

10 Jan 2024

UP Manila, UP Open University

Researchers build a smart computer system that can accurately predict issues when two drugs are taken together

Drug-drug interactions (DDIs) can be a serious problem in healthcare, causing around 30% of unexpected and dangerous medication issues. Over...

Read More

10 Jan 2024

UP Diliman

There is a clear correlation between planktonic foraminifera fluxes and seasonally changing environmental conditions

Planktonic foraminifera are calcareous zooplankton inhibiting the water column. They are sensitive to hydrographic stresses, and variations in their population...

Read More

09 Jan 2024

UP Diliman, UP Los Baños, UP System

Replacing gravel with coconut shell aggregates does not make concrete harder and less prone to corrosion

Utilizing coconut shell aggregates (CSA) in concrete benefits agricultural waste management and reduces the demand for mineral resources. Several studies...

Read More

05 Jan 2024

UP Diliman

Social media fueled widespread dissemination of disinformation narratives during the 2022 Philippine national election campaign period

Social media played a significant role in the 2022 Philippine national elections. Using various empirical sources, including an original pre-electoral...

Read More

Research

The development of an efficient and reliable transport system has a direct impact on the economy of the country. One of the ways that we can improve traffic management and monitoring is to make use of CCTVs, analyze the traffic video data and provide useful traffic information for better monitoring and management. This paves the way for what we call Intelligent Transportation Systems (ITS) where sophisticated algorithms such as machine learning and AI agents provides services to model and manage traffic systems, make them safer, more coordinated and makes efficient use of the transportation network. To enable the use of machine learning systems for ITS, there is a need to train the system on huge amounts of traffic data. The training data are labeled video data, the labels contain the necessary information that the machines need to learn, and it traditionally done manually consuming a lot of time and effort to develop.

The active learning framework is based on the idea that an algorithm can be used to determine which data is more useful and more informative to the machine learning system, so that only those need to be labeled manually. The idea is that the machine learning system will identify the video sequences that have low confidence scores, the active learning model will select the most informative of those sequences with low confidence and they will be manually labeled. After labeling those sequences will be included in the training data of the machine learning, improving the performance and also improving the identification of low
confidence sequences and the process repeats itself. After several iterations machine learning system is getting better at automatically labeling the video sequences and the active learning agent facilitates this improvement. The ultimate goal is to have a machine learn to automatically label unlabeled video sequences thus improving the training process as new data is generated.

Based on our small experiment, using an active learning system generates training sequences that are comparable with systems that learn from randomly labeled video sequences. In terms of traffic information such as vehicle tracking systems, the machine learning model trained using active learning frameworks performs slightly better than systems that rely on the same amount of randomly selected training data.

Authors: Adonais Ray Maclang, Miguel Lorenzo Orante, Rennuel Don Salvador, Dale Joshua Del Carmen, Rhandley D. Cajote (Electrical and Electronics Engineering Institute, University of the Philippines Diliman)

Read the paper: https://ieeexplore.ieee.org/document/10322372