Cross-Border E-commerce Talent Recommendation using Custom Attention with Recurrent Neural Network

In the global commerce dynamic landscape, cross-border e-commerce has appeared as critical path for commercial to extend their influence and capitalize on global markets. Cross-border e-commerce refers to selling goods and services online to consumers across national borders by paying attention to the use of e-commerce websites. This research proposed a Custom Attention with Recurrent Neural Network (CA-RNN) for cross border e-commerce talent recommendation Model. The collected dataset is used here and it is preprocessed by min-max normalization. The preprocessed data features are extracted by Term Frequency-Inverse Document Frequency (TF-IDF) and recommended by CA-RNN. The attention mechanism with RNN enables to concentrate on relevant parts in input sequence. The RNN is used to handle time-series data and it maintain data from previous time step which enables to capture long-term dependencies in data sequence. The f1-score, precision, accuracy and recall are used for evaluating CA-RNN performance. The CA-RNN achieves 97.82% f1-score, 97.59% precision, 98.37% accuracy, 97.61% recall when compared to state-of-art methods.