. Environment

• OS: Windows, Mac OS, or Linux
• Languages: C, C++, C#, Java, or Python (any version is ok)

2. Goal: Perform clustering on a given data set by using DBSCAN.

3. Requirements

The program must meet the following requirements: l Execution file name: clustering.exe

• Execute the program with four arguments: input data file name, n, Eps and MinPts
  • Three input data will be provided: input1.txt, input2.txt, input3.txt
  • n: number of clusters for the corresponding input data
  • Eps: maximum radius of the neighborhood
  • MinPts: minimum number of points in an Eps-neighborhood of a given point
  • We suggest that you use the following parameters (n, Eps, MinPts) for each input data
    • For input1.txt, n=8, Eps=15, MinPts=22
    • For input2.txt, n=5, Eps=2, MinPts=7
    • For input3.txt, n=4, Eps=5, MinPts=5 n Example:

Input data file name = input1.txt, n = 8, Eps = 15, MinPts = 22 l File format for an input data

[object_id_1]t[x_coordinate]t[y_coordinate]

[object_id_2]t[x_coordinate]t[y_coordinate]

[object_id_3]t[x_coordinate]t[y_coordinate]

[object_id_4]t[x_coordinate]t[y_coordinate]

• Row: information of an object
  • [object_id_i]: identifier of the ith object
  • [x_coordinate], [y_coordinate]: the location of the corresponding object in the 2-dimensional space
• Example:

Figure 1. An example of an input data.

l Output files

• You must print n output files for each input data
  • (Optional) If your algorithm finds m clusters for an input data and m is greater than n (n = the number of clusters given), you can remove (mn) clusters based on the number of objects within each cluster. In order to remove (mn) clusters, for example, you can select (mn) clusters with the small sizes in ascending order
  • You can remove outlier. In other words, you dont need to include outlier in a specific cluster n File format for the output of input#.txt input#_cluster_0.txt

[object_id]

[object_id]

• input#_cluster_1.txt

[object_id]

[object_id]

• input#_cluster_n-1.txt

[object_id]

[object_id]

• output#_cluster_i.txt should contain all the ids belonging to cluster i that were obtained by using your algorithm
• Supposed to follow the naming scheme for the output file as above