查詢結果分析
相關文獻
- 遊戲音樂設計與玩家情感反應之研究
- 遊戲音效與玩家情感反應之研究
- 流行音樂風格與情感反應研究
- 八位元遊戲音樂與情感反應關聯性之初探
- 以機器學習偵測異常狀態:龍井太陽能發電場案例
- 門診病患對醫院滿意度與重視度之調查以臺灣地區區域醫院為例
- 從海難事件的發生論船員與船舶間互動效能的提昇
- WWW的使用瓶頸因素及使用者區隔探討--以北區大學傳播相關學系學生為例
- Maximum Likelihood Estimation of Factor Analysis Using the ECME Algorithm with Complete and Incomplete Data
- 中醫醫院門診病人選擇醫院之考慮因素及就醫滿意度研究--以臺中市七家中醫醫院為例
頁籤選單縮合
題 名 | 遊戲音樂設計與玩家情感反應之研究=A Study of Players' Affective Response on Video Game Music Design |
---|---|
作 者 | 林佩儒; 楊智傑; | 書刊名 | 商業設計學報 |
卷 期 | 21 2017.12[民106.12] |
頁 次 | 頁154-173 |
分類號 | 440.8 |
關鍵詞 | 情感反應; 遊戲音樂作曲; 因素分析; 支援向量機; 支援向量機遞迴特徵消去; Affective responses; Video game; Music composition; Factor analysis; Support vector machine; Support vector machine recursive feature elimination; |
語 文 | 中文(Chinese) |
中文摘要 | 在電玩遊戲設計產業中,其音樂部分的設計影響力可以讓玩家融入並沉浸於遊戲角色與情境當中,因此如何設計一款有效影響,並滿足玩家玩遊戲時的音樂情感反應模式,是相當重要的遊戲設計課題。本研究以感性工學為基礎,以八位元時代的電玩遊戲音樂做為樣本,針對100首遊戲音樂,以因素分析方法找出五種代表性情感形容詞共15首音樂樣本作為研究,以支援向量機(support vector machine)分類模型用來建立玩家情感反應與遊戲音樂特徵之間的關聯性,並進一步使用支援向量機遞迴特徵消去(support vector machine recursive feature elimination),挑選出最具關鍵性之音樂並展示出每個特徵對於八位元遊戲音樂之重要性。研究結果顯示,情感反應尺度結果歸類出五組代表形容詞,分別是平靜的、冰冷的、奇異的、幽默的以及熱情的,並針對形容詞的音樂樣本進行音訊分析,可看出與三項音樂特徵的情感反應關係,最後透過支援向量機所獲得的預測模型準確率為94%,證明本研究所選用特徵以及特徵分析方式是可行的,而分析結果也提供具體而明確的音樂設計建議。透過本研究所提出的方法,遊戲音樂設計師可以考慮音樂特徵的優先排列順序,針對特定遊戲情感內容或情境,設計出符合玩家需求的遊戲產品。 |
英文摘要 | In the video game industry, it is the game music that enables players to immerse themselves in their roles within the context of the game. It is, therefore, how to collecting affective responses from game player is of crucial importance to music composers wishing to create an appealing game music track. However, the problems inherent in producing game music works have not yet been intensively and systematically investigated. Hence, based on Kansei engineering, this study proposes a novel methodology for get insight and better understanding the players' emotion in playing game that integrates several methods. Factor analysis (FA) identifies representative adjectives for music samples. Three musical features, including tonality, interval, and rhythm, were extracted for analysis. A support vector machine (SVM) classification model established the connection between the musical features and the affective responses of players. Finally, support vector machine recursive feature elimination (SVM-RFE) was employed to identify critical musical features and determine their importance to game music. The emotional response scale results generalized five representative adjectives: calming, cold, fantastic, humorous, and passionate. A signal analysis of the music samples for each adjective revealed their emotional response relationship with the three music features. Finally, the accuracy rate of the prediction model obtained using the support vector machine was 94%, which demonstrates that the features and the feature analysis approach used in this study are feasible. The analysis results also provide clear and specific suggestions regarding music design. Using the proposed approach, game music designers can consider the priority of music features before designing game products that meet game player needs with certain emotional content or scenarios. |
本系統中英文摘要資訊取自各篇刊載內容。