000 | 01953nam a22001817a 4500 | ||
---|---|---|---|
003 | OSt | ||
005 | 20240919160824.0 | ||
008 | 240829b |||||||| |||| 00| 0 eng d | ||
040 |
_aMMSU _cULS |
||
100 | _aBumanglag, Matthew Hence D. | ||
245 |
_aPrediction of moisture content loss in garlic (Allium sativum Linn.) through machine learning / _cMatthew Hence D. Bumanglag, John Christian N. Corpuz |
||
260 |
_aCity of Batac : _bMMSU, _c2024. |
||
300 |
_axv, 114 leaves : _c29 cm |
||
500 | _aUTHESIS (Bachelor of Science in Agricultural and Biosystems Engineering) | ||
504 | _aBibliography: leaves 63-64 | ||
520 | _aMachine learning involves the examination and computational simulation of various manifestations of learning processes. ML has emerged as an effective method for predicting agricultural crop moisture content loss using microclimate parameters such as temperature and relative humidity. The experiments consist of daily observations made twice a day from 9 a.m. to 4 p.m.—using a data logger. The garlic was divided into two treatment groups: inside the structure and shade. These groups had two plots of laid out garlic and two plots of garlic hanged by the stem. The garlic was left to cure for a period of time. Results reveal that the RF model had the highest level of prediction accuracy based on R2 = 0.70, 0.71, 0.29, and 0.24 for Structure Hanged, Structure Laid, Shade Hanged, and Shade Laid, respectively. Also, RSME = 0.60, 0.68, and 0.79, Structure Laid, Shade Hanged, and Shade Laid, respectively. In addition, MSE = 0.61 and 0.46 for Structure Laid and Shade Hanged. Lastly, MAE = 0.41, 0.39, and 0.51 for Structure Laid, Shade Hanged, and Shade Laid, respectively. An accurate prediction of moisture content in field measurement data can mitigate the effects of agricultural, industrial, and urban practices, as well as drying, weather conditions, and storage practices. | ||
942 |
_2lcc _cTHEDIS |
||
999 |
_c23384 _d23384 |