Totolan Data Mining
Totolan Data Mining
Totolan Data Mining
Oleh :
Juendi 1601234133
Kelas :
06 LD11
ABSTRACT
Evolving information technology makes our daily life easier, as well as data transmission.
Over time, each company primarily property insurance company that continues to grow is
certainly generates a large amount of data or a lot of data such as customer or client. The amount
of data that so many would require large storage capacity and require a system that can help
companies in the process of analyzing large amounts of data so that it becomes a useful
information for the company, of course, in the decision-making process. Data Mining is a
solution that can help companies to collect data, classifying the data and analyze large amounts
of data. Because that data mining is required by companies that produce or require large data or
many, generally using data mining companies such as insurance companies and corporate
franchise or retail business that has great scope.
To overcome this, the company must have an application that is able to perform analyzes
and generate the necessary information to the appropriate management . Information technology
in question is Data Mining . Data Mining is a database of companies which has the function to
process and assess the characteristics of any existing data on the company to assist the
management in the decision making process . Assessment of the satisfaction of the client to the
insurance company , type of insurance that is most widely used by the client and the type of
work that clients using any of this insurance , it can be described through data mining so that the
company can find a business plan for the future looks like. Suppose that using the average
insurance is an entrepreneur , then the company can conduct a campaign in places such as the
wholesale centers that its contents entrepreneurs .
1.2 Scope
In this paper , we will discuss several points , namely :
It is important for us to know the difference from the data warehouse with data mining ,
so we can distinguish what is the use of data warehouse and data mining .
An overview of the general architecture of data mining that describe the work flow or
stages of making data mining .
It is important for us to know the phases of the data mining process is like, from the
initial stage to the final stage .
Some examples of the usefulness of the software or data mining software from several
vendors of IT , so that we can determine what data mining software that fits the needs of the
company.
I did a search of books , magazines , articles , or scientific journal in making this paper .
1.4 Goal and Benefit
The purpose of writing this paper :
1. Can determine the type of property insurance that suits your needs ,
According Indrajani (2009, p2), the data is raw facts or observations about physical
phenomena or the usual business transactions and data are objective measures of the attributes or
characteristics of entities such as people, objects, places and events. Representation of fact which
represents an object such as customers, suppliers, employees, students, lecturers, students and so
on are stored in the form of numbers, letters, symbols, text, sounds, images and so on.
Based on the above understanding, it can be concluded that the data is a collection of a
collection of facts obtained from events that occurred and the collection was recorded, classified
and stored in various forms.
According Abdillah (2012 , p1 ) explained that the database is two or more data by
connecting data elements that can be accessed in various ways . Meanwhile, according to
Iskandar and Rangkuti (2008 , p3 ) says that the database is a collection of data related to each
other data .
So the database is data storage that is mutually integrated and interrelated with the
connecting elements which can be accessed in various ways .
2.3 Understanding Data Mining
According to Han et al (2011 , p6 ) explains that data mining is the selection of a data or
by the term gain knowledge from vast amounts of data . In addition , Han et al (2011 , P36 ) to
explain again that data mining is the process of finding an interesting pattern , and knowledge of
large amounts of data . Meanwhile, according to Segall , Guha and Nonis (2008 , p127 ) explains
that data mining is called the discovery of patterns that artifacts on the data and data mining is
the process of analyzing data from different perspectives or views , collects data and specifies the
data that becomes information useful .
Based on the above understanding , it can be concluded that data mining is a process that
analyzes large amounts of data or a lot and make a pattern to be useful information to users.
According to Turban (2007 , P230 ) describes some of the characteristics of data mining ,
namely :
1 ) Data are often buried in a long time are usually many years in the database ,
2 ) Environmental data mining in the form of client-server architecture and system architecture
information ,
3 ) an advanced tool such as additional visualization tool that helps remove the layer of
information buried in files or records ,
4 ) drill data and query tool that supports the control of the user to ask questions and get an
answer as soon as possible ,
5 ) data mining tool which is combined with a spreadsheet and other software tools
development ,
6 ) parallel process for data mining helps in the search for large amounts of data .
2.3.2 Function Data Mining
1 ) Classification
Serves to classify the target class into that category will be selected.
2 ) Clustering
3 ) Association
Works to find the relationship between an attribute or item set based on the number of
items that appear and existing association rule .
Figure 2.3 Product Association
4 ) Regression
Function to search for the prediction of an existing pattern , almost similar to the
classification .
5 ) Forecasting
Serves to forecasting future trends based on the time that has happened before.
6 ) Sequence Analysis
7) Deviation Analysis
Function to search for rare events or events that are unprecedented and events were very
different from the normal state .
BAB 3
DISCUSSION
3.1 Types of Property Insurance
To determine the types of insurance are contained in the field wrote the property ,
following the division of the types of property insurance , according Asuransi Central Asia or
known as the ACA Insurance , that there are four types :
Source: www.aca.co.id
What can be insured ?
1 ) Building ,
2 ) Household Furnishings ,
3 ) Housewares ,
4 ) Machine ,
5 ) Merchandise .
What is covered?
1 ) Fire ,
2 ) Lightning ,
4 ) Landslide ,
5 ) Flood ,
Any individual or business entity that has an interest in the property insured as : owner ,
tenant , bank or financial institution lenders .
3.1.2 Theft Insurance and Demolition
Source : www.aca.co.id
What can be insured?
Each property to be insured shall be given a detailed description, such as: brand, type,
year of manufacture, purchase price, number of units. These details are very necessary, the
number of units. These details are required both by customers and insurers. Why is that? In case
of burglary, the customer is easy to make a claim in the form of goods to the specifications as in
the appendix at the same insurance policy forecasts the losses occurred.
What is covered?
Loss of items due to theft which preceded acts of violence or coercion or followed by the
destruction of the building. In addition, guaranteed also damage to goods or buildings as a result
of this violence.
Damage caused by the actions themselves or family members and Losses can be insured
through fire insurance or insurance of glass; securities, shares, banknotes, coins, documents and
the like, unless explicitly stated in the summary of the policy; war, riots and the like; government
regulations
Company or individual who has insured the building first with a fire insurance policy.
Buildings insured property is located may be residential buildings, offices, kiosks, cafes, garment
factories or other businesses.
Source : www.aca.co.id
1 ) complete protection ,
4 ) Pay the cost of premiums easy to use ATM BCA , Credit Card, Mobile Banking ,
5 ) To provide additional benefits such as architect's fees , cost of firefighters , the cost of
cleaning debris , without any additional premium costs ,
What is covered?
2 ) Riots ,
9 ) Cleaning debris .
Warranty Extension ?
Source : www.aca.co.id
What is covered?
3 ) Loss of money ,
4 ) Losses due to transport of goods ,
Additional Benefits ?
9 ) Non prorate .
According to Olson & Delen (2008 , p10 ) there are several grooves or work stages of making
data mining :
1 ) Business Understanding
Determine the company's business objectives , assess the current business situation , and
the purpose of making data mining .
2 ) Data Understanding
Finding and collecting the data used by considering the requirements of the required data.
3 ) Data Preparation
4 ) Model Building
Making the initial analysis , sharing of training and testing data sets , data mining and
modeling used .
Checking the level of accuracy of the model was made and evaluate it.
6 ) Deployment
Make an application program user interface from the data mining to display to the user of
the results of the model are made .
The first step in the data mining process steps are data cleaning. Its function is to remove
and replace the inconsistent data, it has no value, irrelevant or are writing errors when inputting.
At this stage it does merging data from multiple tables into a new table. Not infrequently
if data necessary for data mining comes only from one table only, but also from some tables and
some files were covers.
At this stage is the stage of the election or selection of relevant data that is currently on
the table, not all data is only appropriate data for analysis to be taken from the table. In this study
after data cleaning phase and the data phase Intergration already done, done selecting attributes
to select relevant data or in accordance with the analysis based on company needs.
At this stage, the data that has been selected at the stage of selection of data to be used for
the manufacture of data mining models. Data mining models is useful to perform the process on
the data to be used for the analysis process generates very useful information on the company
and the resulting information can be used in making a decision.
1) Orange
open source data visualization and analysis for beginners and experts,
Data mining through visual programming or Python scripting, used for bioinformatics
and text mining.
Equipped with features for data analysis.
Figure 3.6 Example of Data Mining Software Orange
Source : www.orange.biolab.si
2) Weka
Weka is a Java -based software language and set of machine algorithms for data mining tasks
. The algorithm can be applied directly to a dataset or called from your own Java code . Weka
contains tools for data pre - processing , classification , regression , clustering , association rules ,
and visualization .
Figure 3.7 Example Weka Data Mining Software
Source : www.cs.waikato.ac.nz
3) R
R is a software for statistical computing and graphics . R can be run on multiple platforms
such as UNIX , Windows and MacOS
Source : www.r-project.org
4) Microsoft Analysis Services
Microsoft analysis services is a data mining software made by Microsoft . This software is
based on SQL Server Analysis Services to build a model of analysis that can be used for
interactive analyze the data , the data reporting and data visualization . SQL Server provides a
comprehensive model to support the right solution .
Source : www.microsoft.com
Oracle Data Mining (ODM) provides data mining functionality as SQL functions in the
Oracle database . ODM enables users to find information hidden in the data . With the ODM can
build and implement predictive models that help companies to target loyal customers , knowing
customer profiles in detail to prevent fraud . ODM models can be incorporated into SQL queries
contained in the application .
Figure 3.10 Example of Software Data Mining Oracle Data Mining
Source : www.oracle.com
BAB 4
CONCLUSION AND RECOMmENDATIONS
4.1 Conclusion
The continued development of information technology in various fields have made our
lives so completely dependent on information technology. As well as sending data to advance
still using mail or post, now in the presence of information technology such as computers,
gadgets, and the internet can help us to do the online data transmission (e-mail) without the
limitations of time and place restrictions. Currently, the development of business in a property
insurance company continues to grow rapidly because of the many customers or clients who use
the services of the property insurance and property insurance is very useful for customers or
clients to protect their assets from a variety of problems that may come up. Of the many
customers or clients must have also increased in the company data, such as customer data or
insurance clients, financial data, the data of insurance premiums and so on. The large number of
data which would require greater storage capacity and requires an information technology that is
able to gather important information from various data and compiled into one data used for the
analysis.
Information technology mean that Data Mining. With the data mining can help
management to find and collect important information from a variety of existing data to do
analysis, such as the management of customer data or retrieve information insurance clients
(address, age, occupation, type of insurance used by the customer, the premium bill The client
and reports of incidents that often occur such as: fire, flood, etc.). From the information can
indicate the results of which will be used for the company's management decision-making
process and what strategies will be used to maintain current customers and new customer targets.
4.2 Recommendations
Here are some suggestions that can be given is :
1.Necessary care and maintenance (maintenance ) on a regular basis , especially in the sphere of
information technology data warehouse , data warehouse as a parent as a storage of important
data that will be used to do data mining process .
2. Perform grouping the data according to the type of data, so that data becomes more orderly
and easily searchable , eg customer data or employee data to place clients scattered , making it
difficult for management to seek such data .
Abdillah, A. S. (2012). Penerapan Cluster Table Pada Basis Data Perpustakaan Online dengan
Oracle 11g. Jurnal IEEE Skripsi Universitas Mercu Buana, 1-8.
Han, Jiawei and Kamber, Micheline. (2011). Data Mining Concepts and Techniques, 3rd edition.
Morgan Kaufman.
Indrajani. (2009). Sistem Basis Data dalam Paket Five In One. Jakarta: Elex Media Komputindo.
Iskandar, A., & Rangkuti, A. H. (2008). Perancangan Sistem Informasi Penjualan Tunai Pada PT.
KLATEN BERCAHAYA. Jurnal Basis Data, ICT Research Center UNAS Vol.3, 1-8.
Kroenke, David, M., Auer, David J. (2010). Database processing : fundamentals, design, and
implementation.(11th edition). Prentice Hall.
MacLennan, J. dkk. (2009). Data Mining with Microsoft SQL SERVER 2008. Wiley: USA.
Segall, Ricard S. dkk. (2008). Data mining of environmental stress tolerances on plants. Journal
Emerald Group Publishing Limited. 37 (1): 127-148.
Turban, Efraim. (2007). Decision Support and Business Intelligence System. New York: Pearson
Education Inc.