Abstract:
Mobile service development involves app popularity prediction based on current behaviors. It benefits app development and targeted investment. Reviews and app interaction affect popularity. Most related studies examine internal factors but not external ones.
Since app interaction promotes or inhibits popularity prediction, external factors matter. Due to its interactivity and dynamicity, app popularity prediction is difficult. 1) interactivity—it is difficult to assess the existence and intensity of interactions; 2) dynamicity—interaction influence, such as promoting or inhibiting popularity, changes over time.
DeePOP, a novel popularity prediction model, uses time-varying hierarchical interactions. This work introduces Hierarchical Interaction Graph to organically characterize app relationships and influence. DeePOP’s prediction model uses internal factors and time-varying hierarchical interactions.
It creates multi-level Recurrent Neural Network modules with attention mechanism and uses their outputs to predict multi-step time series. DeePOP outperforms state-of-the-art prediction methods on a real-world dataset, reducing the Root Mean Square Error (RMSE) to 0.088.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
