Sequence-to-Sequence Learning Based Conversion of Pseudo-code to Source code Using Neural Translation Approach

Pseudo-code refers to an informal means of representing algorithms that do not require the exact syntax a computer programming language. helps developers and researchers represent their using human-readable Generally, can convert pseudo-code into source code different conversion techniques. The efficiency such methods is measured based on converted algorithm’s correctness. Researchers have already explored diverse technologies devise with higher accuracy. This paper proposes novel learning method includes natural language processing-based text preprocessing sequence-to-sequence deep learning-based model trained SPoC dataset. We conducted extensive experiment our designed algorithm descriptive bilingual understudy scoring compared results state-of-the-art Result analysis shows approach more accurate efficient than other existing in terms several performances metrics. Furthermore, proposed outperforms approaches because utilizes two Long-Short-Term-Memory networks might increase