000			02841nam a22002417a 4500
003			FT8762
005			20251105142850.0
008			251105b ||||| |||| 00| 0 eng d
041			_aengtag
050			_aT58 R66 2024
082			_a.
100	1		_aRomero, Filmer Ivan L.; de Vera Jr., George S.; Serrano, Pia Marie P.
245			_aAudcare: Advanced care for children with autism spectrum disorder through with autism spectrum disorder through a mobile application for predicting and managing meltdowns using machine learning algorithms
264		1	_a. _b. _cc2025
300			_bCapstone Project: (Bachelor of Science in Information Technology) - Pamantasan ng Lungsod ng Maynila, 2024
336			_2 text _a text _b text
337			_2unmediated _aunmediated _bunmediated
338			_2volume _avolume _bvolume
505			_aABSTRACT: Children with Autism Spectrum Disorder (ASD) often experience meltdowns, but predicting these events is challenging for guardians due to the complex interplay of behavioral, environmental, and contextual factors. Additionally, there is a significant lack of accessible information regarding effective prevention and management strategies for these meltdowns, leaving guardians unaware of potential triggers and techniques. Compounding this issue is the scarcity of healthcare professionals specializing in ASD, making it difficult for families to find the necessary support and guidance for their children’s needs. This study aims to develop a mobile application that predicts meltdowns in children with ASD and includes a recommendation engine for strategies to prevent or manage these events, as well as a location-based service for finding nearby healthcare providers. Specific objectives include utilizing the CNN-LSTM model (MoViNets: Mobile Video Networks for Efficient Video Recognition) for accurately classifying tantrums versus predicted meltdowns, generating tailored prevention strategies through content-based filtering, and integrating Google Maps to display local healthcare services. The MoViNets model achieved a training accuracy of 82.03% and a validation accuracy of 87.30%, with overall classification accuracy reaching 90%. The recommendation engine demonstrated strong performance, achieving a Precision of 0.92, Mean Reciprocal Rank (MRR) of 0.89, Mean Average Precision (MAP) of 0.87, and 100% coverage, with low error rates (MAE 0.625, RMSE 1.251), indicating that recommended strategies closely aligned withn expert assessment. Evaluation by therapy centers and technology professionals also indicated good functionality (mean 4.33), usability (4.19), reliability (4.14), and security (4.40), resulting in an overall mean rating of 4.27, signifying a well-rounded and effective system for ASD care.
526			_aF
655			_aacademic writing
942			_2lcc _cMS
999			_c37026 _d37026