Web App which will help farmers by recommending crops based on soil pH, temperature,location etc.
ABOUT: This is a predictive system for crop cultivation in certain regions of Maharashtra. Few Machine learning algorithms such as Support Vector Machine (SVM), K-Nearest Neighbour, Naive Bayes are used in predictive system. The model also compares the accuracy obtained from different algorithms where, SVM is found to be beneficial giving accuracy of 86%.
The Packages used were appzcoder, which is used as a CRUD developer with features like admin panel,php-ai which has machine learning algorithms implemented in its library, stevebauman which retrieve a users location from their IP address using external web services, cmfcmf which uses a php API to retrieve and parse global weather data from OpenWeatherMap.org. The system starts by locating the user’s location and forecasting the temperature of the user’s place which becomes one of the input to the system. The second input given is the pH value of the soil and the respective month in which the crop is supposed to be grown. The above three parameters will be used by the system to deliver the best crop fitted to grow which in turn will give the maximum yield.
Crop is a Complete Laravel project which has integrated packages to predict crops.