學術產出-國科會研究計畫

文章檢視/開啟

書目匯出

Google ScholarTM

政大圖書館

引文資訊

TAIR相關學術產出

題名 主動式產品規劃推薦系統之資訊技術進階研究
其他題名 Research of the Advanced Information Techniques for a Proactive Product Planning Recommendation System
作者 陳良弼
貢獻者 資訊科學系
日期 2012
上傳時間 15-四月-2016 11:34:41 (UTC+8)
摘要 推薦系統近年來受到廣泛關注並衍生出多元化的應用,相關推薦技術之研究也為近 年來熱門的研究議題。綜觀現有的推薦技術,以顧客為主軸,按照顧客行為,根據「對 相同物品有興趣的顧客們其偏好在某種程度上應為相近」的假設前提下進行物品推薦, 相關的應用也確實獲得極大成功。然而,以此為出發點之推薦,僅從被動地角度,採被 動式地推薦。 事實上,更具有價值性之推薦模式,應為化被動為主動,從最根本的角度,於企業 產品開發制定初期時,即將顧客偏好納入考量,使預計開發之產品能獲得顧客之喜愛, 成功吸引顧客目光,為企業帶來商業利潤。因此,有別於以顧客為主軸之推薦技術,我 們認為推薦系統的設計可以從企業產品開發制定的角度出發,進行產品開發計畫的決策 推薦,為企業帶來最大的利益。 以產品開發計畫的決策推薦角度來看,主要的核心概念應為產品之於顧客之偏好分 析。唯有掌握顧客之所好,才能掌握市場。然而,精確地掌握顧客偏好進而分析獲取有 用的資訊,實屬不易之舉。顧客偏好分析其背後為許多因素所驅策,如顧客偏好模式、 產品間之競爭關係、產品涵蓋率、設計成本等等因素。如何完整地全面性地掌握上述因 素,也成為精確分析顧客偏好的關鍵所在。 著眼於此,本研究計畫將從企業產品開發制定的角度出發,發展一個主動式企業 產品規劃推薦系統雛型,並擬定兩研究主軸,分別為:「考量天際線性質偏好模式的產 品推薦」、「考量高潛在顧客數涵蓋率之產品推薦」。主要思考各種顧客偏好模型,並結 合現有之資料探勘、資料庫等技術之研究,發展顧客偏好資料分析探勘技術,以期精 確掌握具有高潛在顧客群的產品。本研究計畫將為未來推薦系統相關研究提供可能的 應用模式,而伴隨著本應用系統下所研發的資料分析技術,也將成為未來推薦系統相 關資料分析與資訊應用的基礎,並引領相關研究朝更多元應用層面進行。
Recommendation systems have become extremely common in recent years and are achieving widespread success on the Web. Existing recommendation systems are mainly designed from customers’ perspectives, which focus on making recommendations for customers. In this research proposal, we propose to design a recommendation system aiming at making recommendation for companies when they are making decisions on designing products or planning business strategies. Our recommendation goal is to discover candidate products with high number of potential customers, which help companies to sustain or enhance the competitive nature of the companies by making correct business decisions. The key issue for making the strategy of designing products lies in how the preferences of customers are captured. However, this can be a challenge task, as there are many factors affecting the customer preferences with respect to products. In this research proposal, we will build a recommendation system aiming at providing decision supports for companies to make product designs by considering various customer preferences, preference models, competition models between products, and product coverage issues, develop techniques for precisely handling these factors in making recommendations for company decision supports. The research results will provide possible application models for future recommendation systems, and the developed techniques can also serve as a foundation for future recommendation techniques. Moreover, the results of project will lead the entire area to a new field with more applications and significantly contribute to the technological and academic promotion of our country.
關聯 計畫編號 NSC101-2221-E004-008-MY3
資料類型 report
dc.contributor 資訊科學系
dc.creator (作者) 陳良弼zh_TW
dc.date (日期) 2012
dc.date.accessioned 15-四月-2016 11:34:41 (UTC+8)-
dc.date.available 15-四月-2016 11:34:41 (UTC+8)-
dc.date.issued (上傳時間) 15-四月-2016 11:34:41 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/84757-
dc.description.abstract (摘要) 推薦系統近年來受到廣泛關注並衍生出多元化的應用,相關推薦技術之研究也為近 年來熱門的研究議題。綜觀現有的推薦技術,以顧客為主軸,按照顧客行為,根據「對 相同物品有興趣的顧客們其偏好在某種程度上應為相近」的假設前提下進行物品推薦, 相關的應用也確實獲得極大成功。然而,以此為出發點之推薦,僅從被動地角度,採被 動式地推薦。 事實上,更具有價值性之推薦模式,應為化被動為主動,從最根本的角度,於企業 產品開發制定初期時,即將顧客偏好納入考量,使預計開發之產品能獲得顧客之喜愛, 成功吸引顧客目光,為企業帶來商業利潤。因此,有別於以顧客為主軸之推薦技術,我 們認為推薦系統的設計可以從企業產品開發制定的角度出發,進行產品開發計畫的決策 推薦,為企業帶來最大的利益。 以產品開發計畫的決策推薦角度來看,主要的核心概念應為產品之於顧客之偏好分 析。唯有掌握顧客之所好,才能掌握市場。然而,精確地掌握顧客偏好進而分析獲取有 用的資訊,實屬不易之舉。顧客偏好分析其背後為許多因素所驅策,如顧客偏好模式、 產品間之競爭關係、產品涵蓋率、設計成本等等因素。如何完整地全面性地掌握上述因 素,也成為精確分析顧客偏好的關鍵所在。 著眼於此,本研究計畫將從企業產品開發制定的角度出發,發展一個主動式企業 產品規劃推薦系統雛型,並擬定兩研究主軸,分別為:「考量天際線性質偏好模式的產 品推薦」、「考量高潛在顧客數涵蓋率之產品推薦」。主要思考各種顧客偏好模型,並結 合現有之資料探勘、資料庫等技術之研究,發展顧客偏好資料分析探勘技術,以期精 確掌握具有高潛在顧客群的產品。本研究計畫將為未來推薦系統相關研究提供可能的 應用模式,而伴隨著本應用系統下所研發的資料分析技術,也將成為未來推薦系統相 關資料分析與資訊應用的基礎,並引領相關研究朝更多元應用層面進行。
dc.description.abstract (摘要) Recommendation systems have become extremely common in recent years and are achieving widespread success on the Web. Existing recommendation systems are mainly designed from customers’ perspectives, which focus on making recommendations for customers. In this research proposal, we propose to design a recommendation system aiming at making recommendation for companies when they are making decisions on designing products or planning business strategies. Our recommendation goal is to discover candidate products with high number of potential customers, which help companies to sustain or enhance the competitive nature of the companies by making correct business decisions. The key issue for making the strategy of designing products lies in how the preferences of customers are captured. However, this can be a challenge task, as there are many factors affecting the customer preferences with respect to products. In this research proposal, we will build a recommendation system aiming at providing decision supports for companies to make product designs by considering various customer preferences, preference models, competition models between products, and product coverage issues, develop techniques for precisely handling these factors in making recommendations for company decision supports. The research results will provide possible application models for future recommendation systems, and the developed techniques can also serve as a foundation for future recommendation techniques. Moreover, the results of project will lead the entire area to a new field with more applications and significantly contribute to the technological and academic promotion of our country.
dc.format.extent 1530047 bytes-
dc.format.mimetype application/pdf-
dc.relation (關聯) 計畫編號 NSC101-2221-E004-008-MY3
dc.title (題名) 主動式產品規劃推薦系統之資訊技術進階研究zh_TW
dc.title.alternative (其他題名) Research of the Advanced Information Techniques for a Proactive Product Planning Recommendation System
dc.type (資料類型) report