月旦知識庫
月旦知識庫 會員登入元照網路書店月旦品評家
 
 
  1. 熱門:
首頁 臺灣期刊   法律   公行政治   醫事相關   財經   社會學   教育   其他 大陸期刊   核心   重要期刊 DOI文章
管理資訊計算 本站僅提供期刊文獻檢索。
  【月旦知識庫】是否收錄該篇全文,敬請【登入】查詢為準。
最新【購點活動】


篇名
具情緒辨識之智能聊天機器人
並列篇名
Emotion-aware Chatbot
作者 陳萌智吳俞蓁陳効民盧元欽林昕諺
中文摘要
人類自1760年代工業革命起,致力於以人類與各種生物模型為基礎製造出機械手臂、自動化控制機器人。機器學習經過三次革命,第一波符號邏輯-把人的思維放進電腦、第二波專家系統-把人的所有知識放進電腦,均以失敗收場,而第三次機器學習革命-把人的所有發現放入電腦,給電腦大量資料,卻奇蹟似的發現,電腦可以自己找出規則、然後學習。而今2022年更是出現了劃時代的Chat-GPT對話機器人,不但掀起全球性熱潮,亦是演化出了「黑魔法」,藉由詠唱咒語可以得出對應完整性的,無論是科技業的程式撰寫、金融與行銷計劃書生成、數學物理化學等純論領域的計算過程,甚至占星解卦,可謂應有盡有。儘管Chat-GPT能夠完成各種任務,但有人認為它們仍存在不足,其中一個主要問題是,對話機器人的溝通介面過於制式,難以在交流過程中表現出情感,為了解決此問題,本研究結合智能機器人實踐人臉表情與語音的情緒辨識技術,使Chat-GPT更好理解人類情感並以更富有情感的方式與人類對話,研究結果經實驗後發現,對話過程中使用智能機器人可以透過表情與語音辨識技術分辨使用者的情緒,此外,加入情緒的對話確實會使Chat-GPT產生不同情緒的回應。本研究的成果與發現,可以預見Chat-GPT結合情緒辨識進行自然語言交流,將成為我們未來生活中非常普及化的現象。
英文摘要
Since the Industrial Revolution in the 1760s, humanity has been committed to manufacturing robotic arms and automated control robots based on human and various biological models. Machine learning has gone through three revolutions: the first wave of symbolic logic - putting human thinking into computers, the second wave of expert systems - putting all human knowledge into computers, both of which ended in failure. The third machine learning revolution - putting all human discoveries into computers, giving computers a large amount of data, miraculously discovered that computers can find rules on their own, and then learn. Today, in 2022, there has emerged a groundbreaking Chat-GPT conversational robot, which has not only sparked a global craze but also evolved into ''black magic''. By chanting spells, it can generate corresponding integrity, whether it is programming in the technology industry, financial and marketing plan generation, computational processes in purely theoretical fields such as mathematics, physics, and chemistry, or even astrological divination, it can be said to have everything. Although Chat-GPT can complete various tasks, some people believe that they still have shortcomings. One of the main problems is that the communication interface of conversational robots is too formal, making it difficult to express emotions during communication. To address this issue, this study combines intelligent robot practices with emotion recognition technology for human facial expressions and speech, allowing Chat-GPT to better understand human emotions and converse with humans in a more emotionally rich manner. The research results, after experiments, found that using intelligent robots during conversations can distinguish users' emotions through facial expressions and speech recognition technology. Additionally, incorporating emotions into the conversation does indeed prompt Chat-GPT to generate responses of different emotions. The outcomes and discoveries of this study suggest that the integration of emotion recognition into Chat-GPT for natural language communication is poised to become a widespread phenomenon in our future lives.
起訖頁 86-94
關鍵詞 人工智慧情緒辨識Chat-GPTArtificial IntelligenceEmotion RecognitionChat-GPT
刊名 管理資訊計算  
期數 202409 (13:2期)
出版單位 管理資訊計算編輯委員會
該期刊-上一篇 關聯規則應用於智慧工廠生產績效管理之研究
該期刊-下一篇 AI時代自主健康管理之實踐
 

新書閱讀



最新影音


優惠活動




讀者服務專線:+886-2-23756688 傳真:+886-2-23318496
地址:臺北市館前路28 號 7 樓 客服信箱
Copyright © 元照出版 All rights reserved. 版權所有,禁止轉貼節錄