INTRODUCTION
Dimension Reduction is the process of converting an n-dimensional space problem to a r-dimensional space problem where r < n. It is one of the important preprocessing steps where descriptions of the input features is high dimensional and only some of the features are relevant or significant with respect to the application. For example, in case of diagnosis of cancer, thousands of genes are collected but only some of them are useful in diagnosing the disease. Thus if other features are present then they may cause reduction in accuracy of diagnosis. Hence dimension reduction is one of the useful tools in areas of data mining, pattern recognition and machine learning and it helps to devise efficient algorithms for classification [2]. The main aim of dimension reduction is to search for the optimal set of features which will help to improve the accuracy. Two basic strategies used for this are[2]
Feature Selection Methods: These methods select some of the features from the original dataset based upon some criteria. They discard the redundant or least information carrying features. Some of the techniques in feature selection include the ones based on criteria like dependency, relevance and significance. Advantage of these methods is that they are easily computable and less costly but the disadvantage is that some of the information is lost which may result in reduction of accuracy of the classifier.
Feature Extraction Methods: These methods extract new features using the information contained in the features present in the original dataset. Some of the techniques included in this category are Principal Component Analysis, Independent Component Analysis and Linear Discriminant Analysis. Advantage of these methods is tha...
... middle of paper ...
...n., vol. 52, no.3, pp. 408426, Mar. 2011.
[6] A. Chouchoulas and Q. Shen,Rough set-aided keyword reduction for text categorization, Appl. Artif. Intell.,vol. 15, no. 9, pp. 843873,Oct.01
[7] R. Jensen and Q. Shen,Semantics-preserving dimensionality reduction:Rough and fuzzy rough-based approach, IEEE Trans. Knowl. Data Eng., vol. 16, no. 12, pp.14571471, Dec. 2004.
[8] Q. Hu, D. Yu, J. Liu, and C.Wu,Neighborhood rough set based heterogeneous feature subset selection, Inf. Sci., vol. 178, no. 18, pp. 3577-3594, Sep. 2008.
[9] H. Peng, F. Long, and C. Ding,Feature selection based on mutual information criteria of max-dependency, maxrelevance, and min-redundancy, IEEE Trans. Pattern Anal. Mach. Intell., vol. 27, no. 8, pp. 1226-1238,Aug.2005
[10] http://datam.i2r.a-star.edu.sg/datasets/krbd/
[11] http://archive.ics.uci.edu/ml/
[12] http://www.cs.waikato.ac.nz/ml/weka/
Kay Arthur teaches how to recognize key words and phrases by creating lists, summarizing chapt...
One real world example of a feature search is when people are driving and looking for a place to eat, such as McDonald’s, they look for the color yellow and the shape of the arches to stand out. This happens quickly because the yellow color stands out against the blue sky and the rigid arched shape stands out against the puffy clouds/ smooth sky.
... middle of paper ... ... In Intelligent Data Engineering and Automated Learning–IDEAL 2006 (pp. 1346-1357. Springer Berlin, Heidelberg.
The second model is the LEARN model which was developed by Dr. Elois Ann Berlin and Dr. William Fowkes (1983). It has similar objectives to the previous pneuomnic, which are to provide health care practitioners with a simple pneumonic to improve cross-culture communication. This model should not be viewed as a separate pneumonic to memorize but rather a complimentary method to implement culturally appropriate health care for all patients.
Classification Text documents are arranged into groups of pre-labeled class. Learning schemes learn through training text documents and efficiency of these system is tested by using test text documents. Common algorithms include decision tree learning, naive Bayesian classification, nearest neighbor and neural network. This is called supervised learning.
Data mining is process of computing the data from the large data sets involving methods on to intersection of statistics, machine learning,
Principal Component Analysis (PCA) is a multivariate analysis performed in purpose of reducing the dimensionality of a multivariate data set in order to recognize the shape or pattern of that data set. In other words, PCA is a powerful technique for pattern recognition that attempts to explain the variance of a large set of inter-correlated variables. It indicates the association between variables, thus, reducing the dimensionality of the data set. (Helena et al, 2000; Wunderlin et al, 2001; Singh et al, 2004)
After discussions, a multiple discriminant analysis (MDA), a statistical technique, was chosen. MDA was used primarily to classify and make prediction in problems where the dependent variable was in qualitative form, e.g. bankrupt or non-bankrupt, or a business. The primary advantage of MDA was its ability to sequentially examine individual characteristics.... ... middle of paper ...
Data mining has emerged as an important method to discover useful information, hidden patterns or rules from different types of datasets. Association rule mining is one of the dominating data mining technologies. Association rule mining is a process for finding associations or relations between data items or attributes in large datasets. Association rule is one of the most popular techniques and an important research issue in the area of data mining and knowledge discovery for many different purposes such as data analysis, decision support, patterns or correlations discovery on different types of datasets. Association rule mining has been proven to be a successful technique for extracting useful information from large datasets. Various algorithms or models were developed many of which have been applied in various application domains that include telecommunication networks, market analysis, risk management, inventory control and many others
Normalization is the process of identifying the one best place where each fact belongs, it is being used to minimizinge data redundancy and optimizinge data structure by systematically and properly placing data elements in appropriate g...
Ÿ The information should also be reliable so the user can make informed decisions. Other features are: Ÿ The information should be free of any bias. The information should be understandable.
HAND, D. J., MANNILA, H., & SMYTH, P. (2001).Principles of data mining. Cambridge, Mass, MIT Press.
Jurafsky, D. & Martin, J. H. (2009), Speech and Language Processing: International Version: an Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition, 2nd ed, Pearson Education Inc, Upper Saddle River, New Jersey.
middle of paper ... ... Olckers, Gibbs, Duncan 2007:2. Knowing Dimensions - What is it The knowledge dimension focuses on gaining the relevant information and understanding of healthcare as well as the skills needed to provide it. It also encompasses the “appropriate application” of this knowledge.
Machine learning systems can be categorized according to many different criteria. We will discuss three criteria: Classification on the basis of the underlying learning strategies used, Classification on the basis of the representation of knowledge or skill acquired by the learner and Classification in terms of the application domain of the performance system for which knowledge is acquired.