Scholarlink

<p>Speed dating has become popular for people to meet possible partners in a brief, organized setting. This study investigates the possibility of forecasting the chances of a relationship lasting over the first meeting using machine learning (ML) techniques. To boost prediction accuracy, this research uses the K-Nearest Neighbor Classification (KNNC) method in conjunction with the Brown Bear Optimization (BBO) and Grasshopper Optimization Algorithm (GOA). The KNNC technique is deployed to identify pairs of speed daters using various characteristics derived from their interactions throughout the event. These elements include behavioral signals, speech habits, and demographic data. Using the KNNC technique, the scheme attempts to discover commonalities between couples and forecast the likelihood of relationship continuance. The results of experiments show that integrating optimization methods with the KNNC model leads to potential performance improvements. Nevertheless, a comparison investigation suggests that the KNNC model upgraded with the GOA surpasses the BWO. Furthermore, the study provides concrete proof that the KNNC technique outperforms the KNBW model, which includes the BWO, especially during the testing phase. The KNBB model obtains a training accuracy of 0.932, outperforming the KNGO scheme’s accuracy of 0.886 during the same period. Finally, the findings highlight the effectiveness of ML approaches, notably the KNBB model, in predicting relationships that result from speed-dating sessions.</p>

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

Predicting first date compatibility by using ML to forecast match rates for couples

  • 期刊论文

  • 发表日期：

    2026年3月4日

查看全文
  • Yanhua Bai

摘要

Speed dating has become popular for people to meet possible partners in a brief, organized setting. This study investigates the possibility of forecasting the chances of a relationship lasting over the first meeting using machine learning (ML) techniques. To boost prediction accuracy, this research uses the K-Nearest Neighbor Classification (KNNC) method in conjunction with the Brown Bear Optimization (BBO) and Grasshopper Optimization Algorithm (GOA). The KNNC technique is deployed to identify pairs of speed daters using various characteristics derived from their interactions throughout the event. These elements include behavioral signals, speech habits, and demographic data. Using the KNNC technique, the scheme attempts to discover commonalities between couples and forecast the likelihood of relationship continuance. The results of experiments show that integrating optimization methods with the KNNC model leads to potential performance improvements. Nevertheless, a comparison investigation suggests that the KNNC model upgraded with the GOA surpasses the BWO. Furthermore, the study provides concrete proof that the KNNC technique outperforms the KNBW model, which includes the BWO, especially during the testing phase. The KNBB model obtains a training accuracy of 0.932, outperforming the KNGO scheme’s accuracy of 0.886 during the same period. Finally, the findings highlight the effectiveness of ML approaches, notably the KNBB model, in predicting relationships that result from speed-dating sessions.