2. About me
• Education
• NCU (MIS)、NCCU (CS)
• Work Experience
• Telecom big data Innovation
• AI projects
• Retail marketing technology
• User Group
• TW Spark User Group
• TW Hadoop User Group
• Taiwan Data Engineer Association Director
• Research
• Big Data/ ML/ AIOT/ AI Columnist
2
27. Hard clustering v.s. Soft clustering
• Hard clustering
• Each point is assigned to a one and only one cluster (hard assignment)
• With K-means we try to find K centroids {μ1,…,μK} and the corresponding
• Soft clustering (Fuzzy C-Means Clustering: FCM)
• Each point is assigned to all the clusters with different weights or probabilities
(soft assignment)
27
28. Hard clustering v.s. Soft clustering
• 演算法步驟
• 透過模糊分群的方式,計算每個訓練樣本屬於各群的程度
• 該方法利用了隸屬值 (membership value) 進行模糊化
• 類似機率的概念,每個樣本最終的隸屬值之和為1
• 演算法步驟 (1)
• 訓練樣本 x
• 劃分 c 群
• ci 為該群的中心點
• 某樣本 xj 屬於 ci ,隸屬值表示為 uij
28