Scholarlink

This study compares the effectiveness of a natural language processing (NLP) algorithm with a conventional neural network in predicting mosquito species through wingbeat classification to enhance accuracy. The dataset used in this study is made up of recordings of mosquito wingbeat noises that were sourced from [Kaggle]. The NLP algorithm and conventional neural network are applied to analyze and classify wingbeat patterns for different mosquito species. Sample sizes are determined using [specific method], ensuring statistical power and significance. Results: The results show that when mosquito species are classified using wingbeat patterns, the NLP algorithm produces a far higher accuracy rate (90.3200%) than the conventional neural network (82.6240%). Statistical analysis reveals a notable difference between the two strategies, with a p-value of 0.002 from the independent sample T-test, indicating a significant result. This research confirms the superiority of the NLP algorithm over the conventional neural network in accurately predicting mosquito species based on wingbeat classification.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mosquito Species Wingbeat Classification Prediction Using the Natural Language Processing Algorithm Compared to Conventional Neural Networks for Enhancing the Accuracy

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • N. Isaiah,
  • K. Nagesh

摘要

This study compares the effectiveness of a natural language processing (NLP) algorithm with a conventional neural network in predicting mosquito species through wingbeat classification to enhance accuracy. The dataset used in this study is made up of recordings of mosquito wingbeat noises that were sourced from [Kaggle]. The NLP algorithm and conventional neural network are applied to analyze and classify wingbeat patterns for different mosquito species. Sample sizes are determined using [specific method], ensuring statistical power and significance. Results: The results show that when mosquito species are classified using wingbeat patterns, the NLP algorithm produces a far higher accuracy rate (90.3200%) than the conventional neural network (82.6240%). Statistical analysis reveals a notable difference between the two strategies, with a p-value of 0.002 from the independent sample T-test, indicating a significant result. This research confirms the superiority of the NLP algorithm over the conventional neural network in accurately predicting mosquito species based on wingbeat classification.