dc.contributor.author | Asiimwe, Orla Nerys | |
dc.contributor.author | Babirye, Petrina Tusubira | |
dc.contributor.author | Kivumbi, George Owen | |
dc.contributor.author | Kyebagonza, Jonathan | |
dc.date.accessioned | 2022-06-02T08:45:17Z | |
dc.date.available | 2022-06-02T08:45:17Z | |
dc.date.issued | 2022-04-27 | |
dc.identifier.citation | Asiimwe O. N., Babirye, P. T., Kivumbi G. O., & Kyebagonza, J. (2022). Agri-predict mobile application. (Unpublished Undergraduate Dissertation). Makerere University, Kampala, Uganda. | en_US |
dc.identifier.uri | http://hdl.handle.net/20.500.12281/12976 | |
dc.description | A project report submitted to the School of Computing and Informatics Technology in partial fulfillment of the requirements for the award of the degree of Bachelor of Science in Computer Science of Makerere University. | en_US |
dc.description.abstract | Fluctuation of maize prices in Uganda is a common problem. Maize farmers are often emotionally and financially strained when they invest a lot into farming and after harvesting, the maize prices drop significantly. This strain could be eased if there was a means for farmers as well as other stakeholders to have an idea or approximation of what the prices could be during the next harvest season. This research project conducted between March 2021 and December 2021 introduces Agri-predict mobile application, a solution to this problem. Agri-predict is a react native mobile application, that uses machine-learning forecasting techniques by combing sklearn library sktime with regression to forecast the possible maize prices for the upcoming season ahead of time. This gives stakeholders an idea of how to plan for the next season, for example, maize farmers can know whether to sow more or less grain and in addition, it brings emotional comfort by removing price uncertainty.
The machine-learning model considers its independent variables as maize quantity in terms of production and precipitation levels. It also considers maize prices of the harvest season as the dependent variable.It should, however, be noted that maize prices in Uganda are affected by many more factors but only the above mentioned were taken into consideration for this study.
This report highlights similar studies in agriculture price forecasting conducted in other parts of the world under the Literature Review. It also provides a systematic description of methods that were applied during the study in terms of the techniques employed in data collection, analysis, design, implementation, testing, and validation. The report vividly describes how the machine-learning model was built, trained and tested to ensure accuracy.Finally, it presents a discussion on the results of all the above activities. | en_US |
dc.description.sponsorship | Government of Uganda. | en_US |
dc.language.iso | en | en_US |
dc.publisher | Makerere University | en_US |
dc.subject | agricultural computer application | en_US |
dc.subject | Maize production | en_US |
dc.subject | prediction | en_US |
dc.subject | mobile application | en_US |
dc.title | Agri-Predict mobile application | en_US |
dc.type | Thesis | en_US |