Linli, Peng
Unknown Affiliation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

A Cross-Language Information Retrieval Method Based on Multi-Task Learning Linli, Peng
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 4 No. 3 (2024): MALCOM July 2024
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v4i3.1384

Abstract

This study introduces a novel Cross-Language Information Retrieval (CLIR) method employing multi-task learning and soft parameter sharing to enhance neural retrieval models' feature extraction across languages. The approach integrates an interaction-based neural retrieval model with a semantic-based text classification model, exchanging hidden vectors for richer feature representation. Experimental results across four language pairs—English-Chinese, English-Arabic, English-French, and English-German—demonstrate significant performance improvements. The proposed method achieved the highest Mean Average Precision (MAP) scores: 0.419 for EN-ZH, 0.403 for EN-AR, 0.427 for EN-FR, and 0.441 for EN-DE, surpassing other models like BM25, BPNRM, KNRM, KNRM-Trans, and KNRM-Embed. This research underscores the potential of multi-task learning for CLIR, showcasing improved retrieval performance through semantic information and knowledge transfer.
Tomato Pest and Disease Identification Based on Improved Deep Residual Network and Transfer Learning Linli, Peng; Sen, Tjong Wan; Fahmi, Hasanul; Roestam, Rusdianto
Indonesian Journal of Artificial Intelligence and Data Mining Vol 8, No 1 (2025): March 2025
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v8i1.34038

Abstract

Tomatoes are a vital global crop, but their yield can be severely impacted by various diseases like leaf mold and spotted wilt. Early and accurate diagnosis of these diseases is crucial for implementing timely treatments, thereby reducing crop loss. Traditional manual diagnosis often suffers from low accuracy, high costs, and time consumption. To address these issues, this study introduces a method for identifying tomato pests and diseases using an improved residual network and transfer learning. A dataset comprising images of seven common tomato diseases and healthy leaves was created. This study introduces an improved residual network and transfer learning method to accurately identify tomato pests and diseases. The enhanced ResNet50 model, with an attention mechanism and focal loss, achieved 98.10% recognition accuracy. This research not only facilitates early disease detection, reducing crop loss but also minimizes pesticide use, thereby enhancing environmental sustainability and agricultural productivity worldwide.