Differenze

Queste sono le differenze tra la revisione selezionata e la versione attuale della pagina.

--- dm:start:clustering [18/12/2012 alle 14:09 (12 anni fa)] – creata Fosca Giannotti
+++ dm:start:clustering [18/12/2012 alle 14:20 (12 anni fa)] (versione attuale) – [Guidelines for the homework on clustering] Fosca Giannotti
@@ Linea 5: / Linea 5: @@
   * **Data Understanding: useful as a preliminary step to capture some data property that can help the clustering analysis (8 points)**
-       * Distribution data analysis and suitable transformation of variables
+       * Distribution analysis and suitable transformation of variables
        * Elimination of redundant variables by correlation analysis
   * **Clustering Analysis by K-means: (15 points)**
        * Identification of the best value of k
-       * Characterization of the obtained clusters by using both analysis of the k centroids and comparison of the distribution of variables within the clusters and in the whole dataset
+       * Characterization of the obtained clusters by using both analysis of the k centroids and comparison of the distribution of variables within the clusters and that in the whole dataset
   * **Analysis by density-based clustering (7 points)**
@@ Linea 18: / Linea 18: @@
   * **Analysis by hierarchical clustering (Optional - 3 points)**
     * Analysis to be performed on a sampling of the data for scalability reasons
+====== Description of the variables ======
+For each car driver we observe the following quantities, measured over a certain time window of mobile activity:
+  Length = total traveled distance (m.)
+  Duration = total time spent driving (sec.)
+  Count = number of different trips
+  Phighway = distance traveled on highways (m.)
+  Pcity = distance traveled inside cities (m.)
+  Length_arc_crowded = distance traveled on 20% most crowded roads (m.)
+  Pnight = distance traveled at night time (m.)
+  Pover = distance traveled over speed limit (m.)
+  Profile = number of systematic trips, e.g., work-home
+  Radius_g = radius of gyration: sparsity of location from the center of mass of the driver (mean position)
+  Radius_g_L1 = radius of gyration w.r.t. L1: sparsity of location from the driver's most frequent location (e.g., home)
+  Avg_Dist_L1 = average distance from L1:  average distance from the driver's most frequent location (e.g., home)
+  TimeL1L2 = % time spent at locations L1 and L2 (most and second most preferred locations)
+  EntropyArc = entropy on road segment frequencies, measures the diversity of roads traveled
+  EntropyLocation = entropy on location frequencies, measures the diversity of places visited
+  EntropyTime = entropy on hours of the day, measures the diversity of daily patterns
+Notice that there are no missing values in the dataset, hence "0"s are actual "0"s, NOT missing values.