Analysis of Wordle Game Mechanism Based on LSTM and MLP
Yixi Zhou, Hanyang Cao, Xuanbo Jia · Highlights in Science Engineering and Technology · 2023
This paper uses data analysis, deep learning and natural language processing techniques to study the difficulty and result distribution of word guessing based on the Wordle game mechanism. First, a three-layer LSTM model is established to predict the number of reported results, and the correlation between word factors and guessing difficulty is analyzed. Next, a multi-sequence LSTM model combined with NLP models is established to predict the result distribution of specific words at specific times. Finally, an MLP classification model is built to classify the difficulty of words.