近年來，民意調查有廣泛使用在預測競選期間選情動向的趨勢，試將問卷設計為探討選民們投票理念與投票行為意圖，以預測選民投票意向。但現實社會裡，並不能以二元邏輯的觀念來劃分人類的思想理念與行為意圖，因為人類的思維、決策、行為中都包含模糊不確定性與非定量化的特質，即集合中元素的隸屬度假若是模糊不確定的，非完全屬於此集合又非完全不屬於此集合，則二元邏輯觀念可能無法詳盡地探訴事件本身所具代表的真相，將會扭曲它背後隱藏的因素意義，故欲應用多元邏輯更精確地描述人類不確定性的模糊特質。 本論文將職務表現、政治形象、候選政見、所屬政黨列為預測台北市長選舉的探究因素，應用模糊統計分析傳統問卷，認為每一位選民將因本身的獨特因素特徵，各視影響選情的論域有著不盡相同重要程度，故考慮個體模糊權重、總體模糊權重、模糊權重數的選定，評定三位台北候選人將各具有不同程度的隸屬函數，利用隸屬度與決定投票意向之間的微妙關係，以最大隸屬度預測選民最終決定的投票意向，將對複雜的選舉情勢作更適合的預測估計。 / Public opinion survey has played an important role and has been utilized in forcasting an election campaign. Trying to get an unbiased and reliable forcast that best explain voters behavior, researchers have put a tremendous effort in designing the survey questions trying to capture the true essence of voters intention and action. In this paper, we apply logic of fuzzy theory to extract the voters inclination to vote for any candidate in the event of uncertainty surrounding the voters that may affect his or her voting decision. This paper has taken into account of candidate performance, public image, political views and political party as discussible factors that many have inflected the process of Taipei mayoral election. Furthermore, we use individual fuzzy weight, total fuzzy weight and fuzzy weight to assign the membership function to the candidates. The candidate who has obtained the maximum membership function based on the above categories is likely to get the vote from the uncertain votes.

模糊理論開始於 1960 年代中期,關於這方面的研究與發展均已獲得相當不錯的成果.其中尤以在群落分析應用上的專題研究更是廣泛.Bezdek 提出的模糊分類演算法,乃根據 Dunn 的C平均法所作的一改良方法.但仍有其缺點,例如,未考慮權重且以靜態資料為主. 有鑑於此,本研究對 Bezdek 之方法加以改進推廣,提出加權模糊分類法.對於評價因素為多變量時,應加入模糊權重的考量.此外更結合時間因素,使準則函數成為動態的模式,將傳統的模糊分類法由靜態資料轉為動態資料形式,以反映真實 的情況. / Research on the theory of fuzzy sets has been growing steadily since itsinception during the mid-1960s. The literature especially dealing with fuzzycluster analysis is quite extensive. But the research on FCM still has somedisadvantages. For instance, the