Browse Articles

Optimizing data augmentation to improve machine learning accuracy on endemic frog calls

Anand et al. | Mar 09, 2025

Optimizing data augmentation to improve machine learning accuracy on endemic frog calls
Image credit: Anand and Sampath 2025

The mountain chain of the Western Ghats on the Indian peninsula, a UNESCO World Heritage site, is home to about 200 frog species, 89 of which are endemic. Distinctive to each frog species, their vocalizations can be used for species recognition. Manually surveying frogs at night during the rain in elephant and big cat forests is difficult, so being able to autonomously record ambient soundscapes and identify species is essential. An effective machine learning (ML) species classifier requires substantial training data from this area. The goal of this study was to assess data augmentation techniques on a dataset of frog vocalizations from this region, which has a minimal number of audio recordings per species. Consequently, enhancing an ML model’s performance with limited data is necessary. We analyzed the effects of four data augmentation techniques (Time Shifting, Noise Injection, Spectral Augmentation, and Test-Time Augmentation) individually and their combined effect on the frog vocalization data and the public environmental sounds dataset (ESC-50). The effect of combined data augmentation techniques improved the model's relative accuracy as the size of the dataset decreased. The combination of all four techniques improved the ML model’s classification accuracy on the frog calls dataset by 94%. This study established a data augmentation approach to maximize the classification accuracy with sparse data of frog call recordings, thereby creating a possibility to build a real-world automated field frog species identifier system. Such a system can significantly help in the conservation of frog species in this vital biodiversity hotspot.

Read More...

Impact of hog farming on water quality of aquatic environments in North Carolina

Kancharla et al. | Aug 08, 2023

Impact of hog farming on water quality of aquatic environments in North Carolina

This study collected samples from water bodies near hog farms and an aquatic environment not near a hog farm. It was hypothesized that water bodies near the hog farms would have lower water quality with higher turbidity, total dissolved solids (TDS), and pH than the water body not in proximity to a hog farm because of water contamination with hog waste. Results showed that the turbidity was 4–6 times higher, TDS was 1.5–2 times higher, and pH was 3 units higher in the 2 experimental locations compared to the control location. This study and its findings are important for understanding the impact of hog farming on the proximal water bodies.

Read More...

Impact of light intensity and electrolyte volume on performance of photo-electrochemical (PEC) solar cell

Patel et al. | Mar 14, 2022

Impact of light intensity and electrolyte volume on performance of photo-electrochemical (PEC) solar cell

Here, seeking to develop more efficient solar cells, the authors investigated photo-electrochemical (PEC) solar cells, specifically molybdenum diselenide (MoSe2) based on its high resistance to corrosion. They found that the percentage efficiency of these PEC solar cells was proportional to light intensity–0.9 and that performance was positively influenced by increasing the electrolyte volume. They suggest that studies such as these can lead to new insight into reaction-based solar cells.

Read More...

Impact of Population Density and Elevation on Tuberculosis Spread and Transmission in Maharashtra, India

Rao et al. | Nov 07, 2021

Impact of Population Density and Elevation on Tuberculosis Spread and Transmission in Maharashtra, India

India accounts for over 2.4 million recorded cases of tuberculosis, about 26% of the world’s cases. This research ascertained the bearing of both the population density and the average elevation above mean sea level (MSL) on the number of cases of TB recorded by the districts in Maharashtra, India. The results found a strong positive correlation between the number of TB cases per thousand people and the population density and a strong negative correlation between the number of TB cases per thousand people and the average elevation above MSL.

Read More...