Woofwoof : a machine learning-based prediction system using a voting classifier algorithm for early detection of common dog diseases / Khryssha Marie N. Andres, Quenzzy J. Navelgas, Althea M. Sanglang, and John Mark M. Solis.
Material type:
TextEAC Cavite : School of Engineering, 2026Description: xi, 266 pages ; 28 cmContent type: - text
- unmediated
- volume
- T CS A54 2026
| Item type | Current library | Call number | Status | Barcode | |
|---|---|---|---|---|---|
| THESIS | EAC Library Cavite EAC Cavite-Thesis Section | T CS A54 2026 (Browse shelf(Opens below)) | Available | ML-TH2843 |
Developmental research
Mixed methods research
Likert scale
Structured interview
CRISP-DM method
Includes bibliographical references
WoofWoof is a web-based machine learning application that aids dog owners in the early identification of 14 diseases among their dogs using a symptom-based questionnaire interface. It uses an ensemble of algorithms known as the Voting Classifier, comprising the Support Vector Classifier, Gaussian Naïve Bayes, and K-Nearest Neighbors, built using 75,000 data records of synthetic yet veterinarian-confirmed instances across 52 predictor variables. The algorithm used for the system was selected based on an experimentation process using k-fold cross-validation. The quality evaluation conducted among licensed veterinarians, IT professionals, and dog owners gave an Excellent overall average score of 3.79 out of 4.00. Finally, Woof Woof is openly deployed with an animal welfare organization called Strays Worth Saving in Batangas, Philippines to contribute to One Health and the Sustainable Development Goals of the UN
There are no comments on this title.