年報中富含眾多資訊，其中包含財務性資訊與文字性資訊，財務性資訊之分析方法已相當成熟，而文字性資訊受限於格式及檔案類型，而降低投資人使用或分析此類資訊之效率。管理階層討論與分析(Management’s Discussion & Analysis of Financial Condition and Results of Operations，以下簡稱MD&A)係管理階層傳達其經營決策觀點予投資人之媒介，投資人可透過閱讀MD&A取得更多資訊，過去學者之研究亦證實該項目內之文字性資訊有其重要性，由於文字性資訊缺乏通用之分類架構，因此投資人需耗費較多時間與成本分析該資訊。本研究自美國科技業上市公司，隨機選取40家企業2012年之年報作為樣本資料，藉由文字探勘技術，運用TFIDF將MD&A文字性內容分類至EBRC針對MD&A所發布之分類架構，建立分類閱讀器，使投資人可利用透過系統分類並彙整之文句，迅速取得所需之文字性資訊，以協助使用者有效率地閱讀這些非結構化之文字資訊，藉以減少資料蒐集之時間，提升文字性資訊之可使用性。 / Annual reports are rich in information, which contains financial information and textual information. While the approach of analyzing financial information is common, textual information is confined by its format or the file type it is stored, thus decreasing the efficiency of analyzing this sort of information. Management’s Discussion & Analysis of Financial Condition and Results of Operations (MD&A) is the vehicle for investor to share the sight of managements’ decision making consideration, through reading MD&A investor could obtain more information. According to past researches, textual information is of importance. Due to the lack of a common framework, investors would consume more time and cost to analyze textual information. This research randomly selected 40 samples from publicly traded technology firms of the United-States. Utilizing text-mining technology and TFIDF, classify textual information of MD&A into the framework EBRC established, developing a classified reader for MD&A. To assist investors read non-constructed textual information efficiently and reduce the time of information gathering, thereby enhancing the usability of textual information.
| Identifer | oai:union.ndltd.org:CHENGCHI/G0100353057 |
| Creators | 吳詩婷, Wu, Shih Ting |
| Publisher | 國立政治大學 |
| Source Sets | National Chengchi University Libraries |
| Language | 中文 |
| Detected Language | English |
| Type | text |
| Rights | Copyright © nccu library on behalf of the copyright holders |
Page generated in 0.0021 seconds