This assignment aims to achieve the following general learning objectives:
To further investigate the topics of the end-to-end process of data mining, data preparation, exploratory data analysis, self-organising maps, and clustering;
To provide an opportunity to conduct research on the above-mentioned topics;
To provide an opportunity to demonstrate insight into the above-mentioned
2 Plagiarism Policy
The Department of Computer Science considers plagiarism to be a serious offence. Disciplinary action will be taken against students who commit plagiarism. Plagiarism includes copying someone else’s work without consent, copying a friend’s work (even with consent) and copying material from the Internet. Copying will not be tolerated in this module.
For a formal definition of plagiarism, the student is referred to http://www.ais.up.ac.za/plagiarism/ index.htm (from the main page of the University of Pretoria site, follow the Library quick link, and then click the Plagiarism link under the Services menu). If you have any questions regarding this please consult the lecturer to avoid any misunderstanding.
Note that all assignments submitted for this module implicitly agree to this plagiarism policy, and declare that the submitted work is the student’s own work. Assignments may be checked using the Turnitin system. After plagiarism checking, assignments will not be permanently stored on the Turnitin database.
Question 1: Introductory Concepts
Give examples of real-world implementations of both a temporal database and a time series database. Provide one example of a temporal database, and one example of a time series database. Your examples must illustrate the differences between the two databases by focusing on what values are stored in
Imagine that you have to implement a neural network, and are told to do so using Argue
against this decision by providing two reasons why MapReduce is not well suited for this implementation.
Question 2: Data Preparation
One approach to deal with missing values is to replace missing values for a tuple with the attribute’s mean over the class of the tuple. Suggest two drawbacks associated with this approach, assuming that such class information exists for each tuple in the data
Identify one disadvantage associated with noise smoothing by means of equiwidth
Describe three drawbacks that are unique to the integration of hard copy data into a consolidated
Question 3: Exploratory Data Analysis
Identify one general problem associated with all exploratory data analysis
Briefly explain why scaling is so important for most data visualisation
Give an example of a real-world situation in which a cumulative histogram should be used, rather than
an ordinary histogram.
Consider Chernoff faces used as glyph Identify one advantage associated with this type
Question 4: Self-Organising Maps
Briefly explain how a self-organising map can still be trained in the presence of missing attribute
Self-organising maps exhibit a phenomenon whereby certain neurons are identified as interpolating units (also known as interpolating neurons). Briefly explain why interpolating units are useful in Self- organising map based visualisations, such as U-matrices.
Consider the SIG* Suggest why the process that adds differentiating conditions to the rule
set is likely to produce very large and complex rule sets.
Question 5: Clustering
Briefly explain why Manhattan distance is more robust to noise and outliers than Euclidean distance
One problem associated with scatter plots is that they only compare two attribute values, resulting in many scatter plots being generated if the data set being analysed has even a moderate number of attributes. Suggest a way in which a clustering algorithm could be used to organise a large number of scatter plots related to a single data set, so that it is easier to detect correlations between attributes
DescriptionIn this final assignment, the students will demonstrate their ability to apply two majorconstructs of the C programming language – Fu
Path finding involves finding a path from A to B. Typically we want the path to have certain properties,such as being the shortest or to avoid going t
Develop a program to emulate a purchase transaction at a retail store. Thisprogram will have two classes, a LineItem class and a Transaction class. Th
1 Project 1 Introduction - the SeaPort Project series For this set of projects for the course, we wish to simulate some of the aspects of a number of
1 Project 2 Introduction - the SeaPort Project series For this set of projects for the course, we wish to simulate some of the aspects of a number of