AaronLim/SHDL_Dataset

该数据集专为微调大型语言模型（如GPT、Mistral和OpenELM）而设计，用于处理与多媒体大学（MMU）和Siti Hasmah数字图书馆相关的用户交互。数据集的目标是促进领域适应，使机构能够创建定制的虚拟助手。数据集包含26872个问题/答案对，涵盖了5个高级类别和15个特定意图。每个条目包括标志、指令、类别、意图和响应字段。数据集还支持多种语言现象，以便进行定制。

This dataset is specifically designed to fine-tune Large Language Models (LLMs) like GPT, Mistral, and OpenELM for tasks related to Multimedia University (MMU) and the Siti Hasmah Digital Library. It contains 26,872 question/answer pairs across 5 high-level categories and 15 specific intents, such as library access information, opening hours, admission applications, course fee queries, and scholarship applications. Each entry includes fields like flags, instruction, category, intent, and response, supporting various linguistic phenomena for customization to train virtual assistants suitable for diverse user demographics.