Proponent/Claimant
Jasten Keneth D. Treceñe
Abstract
MTB-MLE teachers encountered several challenges and used similar strategies in teaching. Some of the challenges encountered by the teachers in MTB-MLE are lack of materials written in Mother Tongue and lack of vocabulary. Teachers use strategies such as translating instructional materials to mother tongue. Therefore, the researchers conducted a research on the aforementioned problem to develop a Convolutional Neural Network algorithm classification model and applied its concept on the development of an android-based application that will classify and recognize local vegetables that will help as additional instructional material for MTB-MLE Teachers. Multimethod research approach was utilized in the study. The researchers also employed phenomenological inquiry and data saturation criterions. Thematic Analysis was used to analyze qualitative data. The method of analysis were used to help the researchers move from a broad reading of the data toward discovering patterns and framing global themes. The experimentations are performed by using 449 vegetable images. The data sets were divided into 3 classes, potato (Solanum tuberosum) with 125 images, tomato (Lycopersicum esculentum) with 163 images and bitter gourd (Momordica charantia L. Amargoso) with 161 images, respectively. Each image is 300 by 300 pixels in dimensions and in JPG format. Based on the experiment conducted the model gained an average of 92.8 % accuracy. Furthermore, the application would be helpful to the teachers as an additional learning material and can be used for formative tasks.
Name of Research Journal
International Journal of Computer Science Engineering. KEJA Publications
Volume and Issue No.
Volume 8, Issue 2
Date/Year of Publication
2019
Citation
Treceñe, J. K. (2019). Development of An Android Based Real-Time Image Recognition Application Using Convolutional Neural Network Algorithm: Addressing Challenges in Mother Tongue-Based of Multilingual Education. International Journal of Computer Science Engineering. KEJA Publications, 8(02).
