University Data Mining Article Review on Intelligent Systems

This report provides an extensive overview of data mining techniques and their applications within the realm of intelligent systems. It begins by defining the data mining process and its significance in the context of modern information technology, emphasizing its role in extracting valuable insights from large datasets to aid in decision-making. The report then details the iterative steps involved in the data mining process, including data selection, cleaning, integration, transformation, mining, and knowledge representation. A significant portion of the report is dedicated to exploring various data mining techniques and algorithms such as classification (decision tree, Bayesian, SVM), clustering, prediction (regression), association rule, and neural networks. Each technique is explained with its sub-parts and applications. Furthermore, the report categorizes different types of data mining systems based on data types, data models, and mining techniques. The report also addresses crucial issues related to data mining, including security, user interface complexities, and performance limitations. The report concludes by underscoring the importance of data mining in prediction and decision-making processes, while also acknowledging the ongoing advancements and challenges in the field.

Running head: INTELLIGENT SYSTEM OF ANALYTICS
ARTICLE REVIEW ON DATA MINING
[Enter Name of Student:]
[Enter Name of University:]
[Author Note:]

INTELLIGENT SYSTEM OF ANALYTICS
EXECUTIVE SUMMARY
The data mining technique and the various tools used for it are described in the report and
various algorithms such as clustering, neural networks and classification are also described. The
issues related in the process of mining are also highlighted in the report.
1
Name
Student ID

INTELLIGENT SYSTEM OF ANALYTICS
Table of Contents
Introduction......................................................................................................................................3
Discussion........................................................................................................................................3
Steps of the mining process.........................................................................................................3
Data Mining Techniques and various Algorithms.......................................................................4
Different categories of the data mining system...........................................................................6
Various issues related with technique of data mining.................................................................6
Conclusion.......................................................................................................................................7
References........................................................................................................................................8
2
Name
Student ID

INTELLIGENT SYSTEM OF ANALYTICS
Introduction
In the modern world of information technology the amount of storage systems and the
data which is stored in them are increasing in a huge manner and thus in order to maintain and
analyze these data to help in the decision making process is very much important. The data
mining process refers to the retrieval of unknown and useful data from the databases in order to
analyze them critically to help in the future decision making processes. The knowledge discovery
process can also be related and connected with the data mining process and technique [1]. The
data mining has certain different approaches and algorithms in terms of analyzing different kinds
of data. The report emphasizes on the various techniques for data mining and approaches used in
the various databases and how they help in the process of decision making. The data mining
process is related to the data present in various data warehouse, relational databases and spatial
databases.
Discussion
Steps of the mining process
The process of mining of data consists of some iterative processes which is useful for
knowledge discovery in the database processes –
Data selection: The most appropriate and relevant data are selected from the various
kinds of databases.
Data cleaning: The data which are irrelevant form the collected data are removed in this
phase of the data mining.
Data integration: The integration phase refers to the act of combining data into a
common source.
Data transformation: The collected data from the collection phase is transformed into
the appropriate form in the transformation phase for the appropriate procedure of mining.
Data mining: The potentially useful patterns from the database are extracted using
various algorithms and techniques [2].
3
Name
Student ID

INTELLIGENT SYSTEM OF ANALYTICS
Knowledge representation: The knowledge phase is associated with the representation
of the different kinds of knowledge gathered and representing it to the user in a visual manner
[4]. The various visualization tools are used so that the users can understand the data mining
results.
Techniques used in Data Mining and various Algorithms
The several techniques and other algorithms that are applicable for the mining process of
data consists of the following algorithms or techniques –
Classification
The classification techniques is one of the most used algorithm in the process of mining
of data and it comprises of certain pre-defined examples which helps in the development of
models for the classification and evaluation of huge amount of data [3]. The accuracy of the
models are determined by the neural based network algorithms or the decision tree algorithms.
The data collected from the databases are analysed using the classification algorithms and the
proposed model given by this is termed as a classifier [9]. The classification model is again sub
divided into five parts –
 Decision tree induction
 Bayesian Classification
 Association based classification
 SVM
 Neural Network
Clustering
The clustering algorithm technique is used for the identification of the different similar
classes of the objects that are created and with the application of the clustering technique the
various dense and sparse regions in the database that is to be analysed, is analysed and various
relations are also made among the attributes of data [8]. However, the process of clustering is
expensive in many aspects for the purpose of distinguishing the classes of object and due to this
reason the clustering is sometimes used as approach for pre-processing selection of subsets of the
4
Name
Student ID

INTELLIGENT SYSTEM OF ANALYTICS
attributes. The clustering algorithm or technique also have certain sub parts which determines the
patterns such as –
 Agglomerative method
 Methods based on model
 Methods based on grid
 Portioning methods
 Methods based on density
Prediction
The prediction of the patterns can be done with the help of the regression technique and it
can be used to relate one or more number of variables which can be dependent or independent
variables. However, the teal world problems are often not predicted by the prediction models
such as stock prices and rate of product failure. The prediction by the regression methods are a
complex methods and used for the future value forecasting and one of the most useful feature is
that similar algorithms can be often used for the classification as well as prediction [6]. The
neural network can also be applied for both the classification models and regression model. The
methods of regression can be further divided into the following sub parts –
 Multivariate Nonlinear Regression
 Regression in linear method
 Multivariate Linear regression
 Regression in non-linear method
Association Rule
The frequent and repetitive data in the large amount of databases are found using the
correlation and association techniques. The correlations in sales in transactional data or other
datasets are usually determined with the use of the association rule present in field of data
mining. The number of frequent items appearing in the database are usually analysed with the
implementation of the association rule [7]. The designing and application of various machine
learning approaches can be developed using the association rule in the field of mining of data.
The association rule can be sub divided into the following sub parts –
5
Name
Student ID

INTELLIGENT SYSTEM OF ANALYTICS
 Multidimensional rule of association
 Multilevel association rule
 Quantitative rule of association
Neural Network
The artificial neural networks are an effective network of interconnected neural used for
the processing of information and these neural networks during its learning phase are usually
adaptive in nature. The various complex relationships between the input and output and the
patterns existing in them are generally determined with the help of the neural network approach.
The neural network can also derive patterns and data from the complicated databases and it also
suited for the continuous output and input channels [10]. The act of pattern identification is best
done using the neural network process and also in the prediction purpose. The neural network
has mainly a sub part known as back propagation which is used in pattern detection.
Different categories of the data mining system
The various kinds of system used for the purpose of mining of data are limited sometimes
to some definite functionalities and other systems might be more comprehensive and versatile
[11]. The system can be sub divided into the following categories –
According to types of data mining: The kinds of data that are mined such as time-series
data, text data and spatial data are categorized on the basis of the process in which they are
mined.
According to data models: The various kinds of data models involved in the purpose of
data mining such as data warehouse and relational database are categorized in this classification.
According to the various mining techniques: The various techniques that are
associated with the data mining such as neural networks, machine learning, visualization and the
data mining system is categorized on the basis of these techniques and it is also regarded as the
most effective classification as because the most effective methods or algorithms are suggested
and classified on the basis of their operation structure.
6
Name
Student ID

INTELLIGENT SYSTEM OF ANALYTICS
Various issues related with technique of data mining
The various issues are related with the process of mining of data as because the technique
is still under development and various complex procedures are related with developing the
techniques related to data mining –
Security issues: The data collection from any database has certain security issues as because in
the process of data collection, huge amounts of data are collected which might be highly
sensitive or private related to a single organization or a group of organizations. The access to
these databases if made without proper permission, may result into illegal activities and also
might violate the privacy policy of the organizations [12]. The data mining technique also widely
spreads the data and information retrieved from the databases and this is done without any
control which also in terms makes security issues.
Issues in user interface: The use of mining of data is only useful and beneficial to the user’s
until they are able to understand the processes and the output. In order to do so, various digital
visualization techniques are used in order to make the user interface better. The results of the
mining of data are represented with the use of visualization techniques so that the users can
understand the results better. However, the various complex results cannot be represented with
the help of visualization and this creates a problem for the users.
Performance issues: The various techniques that are used for the data mining are not often
designed for handling the huge amount of database and thus the efficiency and scalability of
results are hampered which reduces the performance of the algorithms and techniques [5]. Thus
because of the excessive amount of data present in the databases, several performance related
issues are generated in the processes of mining of data.
Conclusion
The process of data mining is one of the most used technique in terms of prediction of
future results and making decisions and it is based on the various techniques and models used for
mining processes. The report also emphasizes on the classification, neural network and clustering
techniques of the mining procedure and thus in can be concluded that mining of data is a
complex and advanced process in the information technology world and with the proper
7
Name
Student ID

INTELLIGENT SYSTEM OF ANALYTICS
development of it various prediction and decision making system can be developed with higher
efficiency. The issues related with the data mining are also described in the above report.
References
[1] P.N. Tan, Introduction to data mining. Pearson Education India, 2018.
[2] I.H. Witten, E. Frank, M.A. Hall and C.J. Pal, Data Mining: Practical machine learning
tools and techniques. Morgan Kaufmann, 2016.
[3] D.T. Larose and C.D. Larose. Discovering knowledge in data: an introduction to data
mining. John Wiley & Sons, 2014.
[4] C.C. Aggarwal, Data mining: the textbook. Springer, 2015.
[5] A.A. Freitas. Data mining and knowledge discovery with evolutionary algorithms. Springer
Science & Business Media, 2013.
[6] N. Jain and V. Srivastava. Data mining techniques: a survey paper. IJRET: International
Journal of Research in Engineering and Technology, 2(11), 2013, pp.2319-1163.
[7] B. Thuraisingham. Data mining: technologies, techniques, tools, and trends. CRC press,
2014.
[8] J.A. Silva, E.R. Faria, R.C. Barros, E.R. Hruschka, A.C. De Carvalho and J. Gama. Data
stream clustering: A survey. ACM Computing Surveys (CSUR), 46(1), 2013, p.13.
[9] G. Kesavaraj and S. Sukumaran. A study on classification techniques in data mining. In
2013 Fourth International Conference on Computing, Communications and Networking
Technologies (ICCCNT). IEEE, July, 2013, pp. 1-7.
[10] S. Poria, E. Cambria and A. Gelbukh. Aspect extraction for opinion mining with a deep
convolutional neural network. Knowledge-Based Systems, 108, 2016, pp.42-49.
8
Name
Student ID

INTELLIGENT SYSTEM OF ANALYTICS
[11] R.L. Grossman, C. Kamath, P. Kegelmeyer, V. Kumar and R. Namburu. Data mining for
scientific and engineering applications (Vol. 2). Springer Science & Business Media, 2013.
[12] D. Braha. Data mining for design and manufacturing: methods and applications (Vol. 3).
Springer Science & Business Media, 2013.
9
Name
Student ID