Hello , I am a graduate student and I would like some help with the following problem related to computational intelligence (soft computing), the image of the problem is attached:
Title: Wine Recognition Data Source: Forina, M. et a PARVUS - An Extendible Package for Data Exploration, Classification and Correlation. Institute of Pharmaceutical and Food Analysis and Technologies, Via Brigata Salemo, Genoa, Italy. Information The se data are the results of a chemical analysis of wines grown in the same region in Italy but derived from three different cultivars. The analysis determined the quantities of 13 constituents found in each of the three tvpes of wines. The initial data set had around 30 variables, but this is the 13 dimensional version. The selected 13 variables are unfortunately unable to be specified for what they are Number of Instances: 178 Attributes .Number of Attributes: 13 All attributes are continuous. No missing attribute value Number of Class: 3 For a given data set above, analyze it by means of each algorithm below. The stopping criteria of algorithm is the error limit of 0.005. For Q1 - Q3, use the Euclidean distance for a distance measure For Q4, use the Gustafson-Kessel algorithm. In order to list the data in each cluster, it's suggested list them with their membership values in the excel sheet. ''By applying Hard C-Means algorithm, classify the given data into 3 clusters. (1) Find each cluster center vi, v2, vg and (2) List the data instance of each cluster by its instance numbers, not by its 13 attribute values: e.g.) Cluster 1 =(z11.115, , } QI, 02. By applying Probabilistic Fuzzy C-Means (FCM) algorithm, classify them into 3 clusters. Use a fuzzifier m 2. (1) Find each of cluster centers v1,2, v and (2) List the data instance of each cluster as a pair of (instance number, a membership value of instance in a cluster), e.g) Ci uc/),) Use the initial fuzry pseudo partition U(0) below. 178 Title: Wine Recognition Data Source: Forina, M. et a PARVUS - An Extendible Package for Data Exploration, Classification and Correlation. Institute of Pharmaceutical and Food Analysis and Technologies, Via Brigata Salemo, Genoa, Italy. Information The se data are the results of a chemical analysis of wines grown in the same region in Italy but derived from three different cultivars. The analysis determined the quantities of 13 constituents found in each of the three tvpes of wines. The initial data set had around 30 variables, but this is the 13 dimensional version. The selected 13 variables are unfortunately unable to be specified for what they are Number of Instances: 178 Attributes .Number of Attributes: 13 All attributes are continuous. No missing attribute value Number of Class: 3 For a given data set above, analyze it by means of each algorithm below. The stopping criteria of algorithm is the error limit of 0.005. For Q1 - Q3, use the Euclidean distance for a distance measure For Q4, use the Gustafson-Kessel algorithm. In order to list the data in each cluster, it's suggested list them with their membership values in the excel sheet. ''By applying Hard C-Means algorithm, classify the given data into 3 clusters. (1) Find each cluster center vi, v2, vg and (2) List the data instance of each cluster by its instance numbers, not by its 13 attribute values: e.g.) Cluster 1 =(z11.115, , } QI, 02. By applying Probabilistic Fuzzy C-Means (FCM) algorithm, classify them into 3 clusters. Use a fuzzifier m 2. (1) Find each of cluster centers v1,2, v and (2) List the data instance of each cluster as a pair of (instance number, a membership value of instance in a cluster), e.g) Ci uc/),) Use the initial fuzry pseudo partition U(0) below. 178
