Table of Contents

..................................................................................................................................1
..................................................................................................................................2
CHAPTER I ..................................................................................................................4
Acknowledge .............................................................................................................5
1.1 Background ......................................................................................................5
1.2 Problem............................................................................................................6
1.3 Objectives.........................................................................................................6
CHAPTER II .................................................................................................................7
Context and Perspective .............................................................................................7
2.1 Market Segmentation .......................................................................................7
2.2 Target customers .......................................................................................8
CHAPTER III .............................................................................................................. 12
Clustering methods .................................................................................................. 12
3.1 Cluster Analysis ........................................................................................ 12
3.2 Hierarchy and Non hierarchy .................................................................... 13
3.3 Data Cluster ................................................................................................... 14
3.2 Cluster pre- processing methods ..................................................................... 15
CHAPTER IV.............................................................................................................. 18
The Data .................................................................................................................. 18
4.1 Content ................................................................................................... 18
4.2 Segmentation .......................................................................................... 19
Chapter V................................................................................................................. 26
Implementation ....................................................................................................... 26
CHAPTER VI.............................................................................................................. 30
Conclusion and Suggestions ...................................................................................... 30
An objective this research is finding how many cluster that probably in the
data, and fixing number of stable cluster. The strength of this research is
dataset that collected isn’t just sample data but its actual data. But for fixing
problems not just a simple method. Kindly for further research should be deep
 in analysis and fixing the visualisation to be a good visualitation for data
intepretation. A good visualization can make a model easy to understand. ........... 31
Bibliography ............................................................................................................ 32
 CHAPTER I
Acknowledge

1.1 Background
Identification of target markets is a necessary step in the planning
and development of marketing strategies. In a situation where the
consumer faces many choices, then the success of product marketing will
be largely determined by the suitability of the product to the needs of
consumers in a particular segment.
The company can not connect with all its customers in a large,
broad, or diverse market. But they can divide that market into a consumer
group or segment with different needs and wants. Then the company
should identify which market segments it can serve effectively. This
decision requires an in-depth understanding of consumer behaviour and
thoughtful strategic thinking. To develop the best marketing plan, managers
must understand what makes each segment unique and different.
To be liked and trusted by consumers, is not an easy thing in the
can therefore must be bought the creation of satisfaction that will be
obtained from the consumer product concerned. This does not mean that
companies only maximize customer satisfaction, but companies must try to
earn profits by way of satisfaction, the company must first learn about what
is purchased, how habits, under what conditions goods and services
purchased by consumers, who will served, how much is needed consume
and why consumers choose a particular product.
In order for the products offered to succeed in the marketplace,
marketing management needs to examine consumer attitudes toward
products from their brands. This attitude causes people to behave
constantly toward the same \. For this reason, attitudes are very difficult to
change and tend to persist in a fixed pattern. Therefore, a company needs
to adapt their products to existing attitudes, rather than trying to change
people's attitudes. To know the attitude of consumers to their products, the
company needs to conduct research on the factors contained in the product
that can encourage the creation of a buying motive.

1.2 Problem
a. What about cluster segmentation?
b. What kind of data that has characteristics to be solved by clustering
method?
c. How the data can be solved by cluster method?
d. How he clusters segmentation formed in many various mineral
water data collections

1.3 Objectives
The objectives in this paper are, as follows:
a. As one of Multivariate Data Course duty.
b. Making this paper aims to describe the market segmentation.
c. Identify Clustering in many various mineral water data collections
 CHAPTER II
Context and Perspective

2.1 Market Segmentation

Market segmentation is a division of buyer groups that have
different needs, characteristics, or different behaviours within a particular
market. Market segmentation can also be interpreted as identifying the
difference analysis of buyers in the market.
Market segmentation according to Philip Kotler and Gary
Armstrong is the division of a market into different buyer groups. Market
segmentation can be intended as a heterogeneous market division into
homogeneous market groups, where each group can be targeted to market
a product according to the needs, wants, or characteristics of the buyers in
the market.
There are several effective segmentation requirements:

• Can be measured (measurable)

The size, purchasing power, and market profile must be measurable at
some level.
• Accessible
Market segments can be reached and served effectively.
• Large enough (substantial)
Market segmentation is large enough or enough to provide profits that can
be served. A segment is a homogeneous group of considerable value to be
served by an appropriate marketing program.
• Differentiated (differentiable)
Differentiable means that segment can be clearly distinguished.
• Can be implemented (actionable)
Actionable means that segment can be reached or served with resources
owned by the company.
The market segment consists of a group of customers who have a
similar set of needs and wants. The job of the marketer is to identify the
segment and decide which segment to target. Segment marketing offers
key benefits over mass marketing. Companies can often design, price,
release, and deliver products or services better and can also tailor
marketing programs and activities to beat competitors' marketing. We can
determine the character of the market segment in various ways. One way is
to identify the preferences segment. Preferred segments consist of:
Homogeneous preferences, there when all consumers have
almost the same preferences, the market does not show a natural segment.
Diffused preferences, diffused preferences have varying
preferences. If there are multiple brands in the marketplace, each wants to
position itself in the competitive space and show real differences to adjust
to differences in consumer preferences.
Group preference, this preference is generated when a natural
market segment emerges from a consumer group with the same
preferences.

2.2 Target customers

Customer Understanding and Grouping
 Understanding customers
Customer is someone who build good relationships with other
people, especially producers in the field of business.
  Combining writers
In order to facilitate the identification of customers, we better
classify the customer as an organization, or an intact institution, and
basically all have two needs, namely physical needs and spiritual
needs. Type of needs in varying varies depending on the situation,
conditions, and purchasing power each.
There are basically two types of customers: external customers and internal
customers.
1. External customers
External customers are people outside the industry who receive a
product (enduser). External customers of every industry is clearly the
general public who receive the industry product tersebut.Beberapa
things in need of external customers :
a. conformity with the need for the product
b. competitive prices
c. quality and reliability
d. timely delivery
e. after sales service.
2. Internal customers
Internal customers are the ones who do the next prises of a job
(NEXT PROCESS) Internal customers are all employees of an industri.Yg in
need of internal customers are:
a. group work and cooperation
b. efficient structures and systems
c. quality work
d. timely delivery
Another customer classification is as follows:
1) Customers in terms of number
a) Induvidu customers
Induvidu customers are customers who shop to meet their own
needs and interests. He does not shop for family, organization or
other people. Induvidu customers are expecting satisfaction in the
product in need right. Each induvidu basically has two needs:
physical needs and spiritual needs. Physical needs are primary
needs, secondary needs, and tertiary needs different from each
induvidu. Spiritual need is the need for satisfaction, for example in
terms of waiters and quality of goods purchased.
b) Family customers
Family customers are customers who shop on behalf of the family
even if the purchase is done by a representative. The customer of
the family will do more volume of purchase or give more
opportunity to the seller.
c) The customer of the organization
The customer of the organization is the customer who made the
purchase to meet the needs of the organization even if it was
purchased in a representative to someone. For that we as a seller is
able to ensure the purchasing volume of the organization's
opportunities whether the needs of weekly, monthly needs or
annual needs.
2) Customers in terms of age
Customers when viewed in terms of age is diverse.semakin increase
the age of a person, the more needs more needs in his need.
3) Customers in terms of sex
Customers when viewed in terms of sex generally consists of men
and women. When viewed from the aspect of population growth,
 the comparison between men and women is more women, and it
shows the target market for more women is more potential.
4) Customers in terms of income
Customer when viewed from the aspect of income generally for the
size of the population of Indonesia is divided into three namely high-
income residents, middle income population, and low-income
residents.
5) Customers in terms of education
The customer when viewed in terms of education level is very
diverse and it is very diverse and it is very influenced by the
differences of taste of each.
6) Customers in terms of time of purchase
Customers who are reviewed from the aspect of the purchase time
vary greatly because it depends on when in need it.
 CHAPTER III

Clustering methods

3.1 Cluster Analysis

Cluster analysis or clustering is the task of grouping a set of
objects in such a way that objects in the same group (called a cluster) are
more similar (in some sense) to each other than to those in other groups
(clusters). It is a main task of exploratory data mining, and a common
technique for statistical data analysis, used in many fields, including
machine learning, pattern recognition, image analysis, information
retrieval, bioinformatics, data compression, and computer graphics.
Cluster analysis itself is not one specific algorithm, but the general
task to be solved. It can be achieved by various algorithms that differ
significantly in their understanding of what constitutes a cluster and how to
efficiently find them. Popular notions of clusters include groups with small
distances between cluster members, dense areas of the data space,
intervals or particular statistical distributions. Clustering can therefore be
formulated as a multi-objective optimization problem. The appropriate
clustering algorithm and parameter settings (including parameters such as
the distance function to use, a density threshold or the number of expected
clusters) depend on the individual data set and intended use of the results.
Cluster analysis as such is not an automatic task, but an iterative process of
knowledge discovery or interactive multi-objective optimization that
involves trial and failure. It is often necessary to modify data pre-processing
and model parameters until the result achieves the desired properties.
 Besides the term clustering, there are a number of terms with
similar meanings, including automatic classification, numerical taxonomy,
botryology (from Greek βότρυς "grape") and typological analysis. The subtle
differences are often in the use of the results: while in data mining, the
resulting groups are the matter of interest, in automatic classification the
resulting discriminative power is of interest.
Cluster analysis was originated in anthropology by Driver and
Kroeber in 1932 and introduced to psychology by Zubin in 1938 and Robert
Tryon in 1939[1][2] and famously used by Cattell beginning in 1943[3] for
trait theory classification in personality psychology. (wikipedia.com, 2018)

3.2 Hierarchy and Non hierarchy

The purpose of cluster analysis is to grouping objects based on
characteristic migrations among the objects. Thus, the characteristics of a
good cluster, which has :
- Internal homogeneity (in clusters), ie seeking between groups.
- External heterogeneity (between clusters), which is the value of one
cluster with other clusters.
The clustering methods (criteria) in the cluster analysis include :
- Hierarchical method, initiate grouping with two or more objects that
have Nearest Photasia. Then forwarded to another object until so on the
cluster will construct a kind of 'tree' where there is a clear (hierarchy) level
between objects, from the like to the least similar. Tools that help to clarify
this hierarchy as "dendogram".

Non-hierarchical methods, starting with determining the number of

clusters first desired (two, three, or the other). After the number of clusters
is determined, then the cluster process done without doing the hierarchy
process. This method can be called "K-means cluster. "K-means cluster is
very effective and efficient at this time to group objects that hide large. K-
mean cluster for more malicious objects of guess.

3.3 Data Cluster

The data used in this study is the price data of each brand of mineral
water in the can from the survey results in each stores / mini
market. The price data of each mineral water consists of various
brands including (Aqua, Vit, Oasis, Prima, Clif, Lemineral, Sanqua,
Club).

The steps taken in this data processing as follows:

1. Install Package
To run a certain R program requires appropriate libraries so
that the instructions / commands typed are recognized by R,
for the current project the libraries used are dplyr, rjava, xlsx,
xlsxjars.
2. Setting Working Directory
Specify the Files storage directory in a folder.
3. Read .xlsx
Read the data to be processed into R
4. Data select
Data is selected to scale.
3.2 Cluster pre- processing methods

Data Processing is a data manipulation to become a more useful

form. Data processing is not only a numerical calculation but also
operations such as data classification and transfer of data from one
place to another by using certain means or formulas. data processing
aims to convert raw data from measurement results into finer data so as
to provide direction for further assessment (Proposal Eksperimen,
t.thn.). The data processing steps consist of three main steps, namely
input, process (processing), and output.

1. Input
In this step the data that will be input is determined Keyword or
Keyword first by determining the amount of data to be taken.
2. Proses
In this step the input data is changed, and usually combined with
other information to produce data in a more usable form. This
processing step usually includes a series of basic processing
operations
3. Output
In this step the results of the previous processing of data that
has been taken and in can by using software Rstudio then
collected. The form of output data depends on the acquisition of
data in the can.

Finally all of our data points are ordered according to their

respective cluster first component, and plotted against the number of
clusters (thus creating the clustergram).
Knime
Knime Analytics Platform is the open source software for creating
data science applications and services. Intuitive, open, and continously
integrating new developments, KNIME makes understanding data and
designing data science workflows and reusable components accessible to
everyone (Open for Information Knime, 2018).

Data Processed by KNIME

Import data in the form of excel file into KNIME used Read Excel
Each column of the table are joined
 Each column are filtered by filter column
 If the processed end, export data with excel format or CSV use write
Excel/CSV.

R Studio
R studio is an integrated development environment (IDE) for R. It
includes a console, syntax-highlighting editor that support direct code
execution, as well as tools for plotting, history, debugging and workspace
management. R studio is available in open source and commercial editions
and runs on the dekstop (Windows, Mac and Linux) or a browser connected
to R studio server or R studio server pro (Debian/Ubuntu, RedHat/CentOS,
and SUSE Linux) (R Studio, 2018).
Data processed by R
 Install R package, which consists of several packages, such as rJava,
dplyr, xlsx, Xslx, rLang, CSV, and others.
 Each library are runned
 Import the data into R
  Coding R by adding some source code and adding source method
used then run all source code, then run all source code so as to
generate plot from data we import

Clustergram in R (a basic function)

After finding out about this method of visualization, I was hunted by

the curiosity to play with it a bit. Therefore, and since I didn’t find any
implementation of the graph in R, I went about writing the code to
implement it.
The code only works for kmeans, but it shows how such a plot can
be produced, and could be later modified so to offer methods that will
connect with different clustering algorithms.
How does the function work: The function I present here gets a
data.frame/matrix with a row for each observation, and the variable
dimensions present in the columns. The function assumes the data is
scaled.
The function then goes about calculating the cluster centers for our
data, for varying number of clusters. For each cluster iteration, the cluster
centers are multiplied by the first loading of the principal components of
the original data. Thus offering a weighted mean of the each cluster center
dimensions that might give a decent representation of that cluster (this
method has the known limitations of using the first component of a PCA for
dimensionality reduction, but I won’t go into that in this post).
 CHAPTER IV

The Data

4.1 Content
How to find the data by way of survey to collect the price of each
brand of mineral water products in each area ie jakarta, bekasi, depok and
bogor then find out the price and mineral water in each ditch area and send
the price on each brand then search know what is interested in buying
mineral water with the brand in each region and the period expired on the
brand of air mineral products of each region.

Table 1: Data Mineral Water.

The following data is the data of brand rank, expiration period and area
rank in every area of Jakarta, Bekasi, depok and bogor. from the data there
are aqua brand with a ranking of one in Jakarta, Depok, Bogor, Bekasi and
second there is a brand of lemineral, the three brands of vit, the four saqua,
the five Oasis, prime, Clif and last there Club. with the expiry period of the
brand is an average.

Table 2: Mineral Water

4.2 Segmentation

Data collection : The following data are collected from various sources, to
be grouped and compared according to each brand.
The following table on the data:
• Brand
Data taken from various brands to make a difference, consisting of brands:
 1. Aqua
2. Vit
3. Oasis
4. Prima
5. Clif
6. Lemenal
7. Sanqua
8. Club
• ML
Data /ml to compare selling price in selling / ml.
• Territory
This data is to compare prices in different regions.

Image of Aqua mineral water data

The picture above is mineral water data with aqua brand, and we can see
the data in every area of Jakarta, Bekasi, depok and bogor with unit of box
with the same quantity and ml turns the price in every area also have same
price.

Image of mineral water data of VIT brand

In the picture above is the data from the brand mineral water VIT, with
units of boxes and the contents of the same box, ml also the same price in
each area are also the same one box.
 Image of Oasis brand mineral water data

From the image data above Oasis brand data is not much different from the
data of other mineral water brands, one box has the same content or
quantity and the same ml and the same price.
 Image of mineral water data of Prima brand

Prima mineral water brands are the same, with the contents of the boxes,
quantity, and ml have the same price in each are.

Clif mineral water brand image

Picture above mineral water data of Clif brand has the contents box and ml
same with same price in every area.
 Drink mineral water Le mineral brand

The mineral water data of Lemineral brand in each area has the same
quantity and ml and the price is the same in every area.

Sanqua brand mineral water image

Sanqua brand mineral water data is not much different from other brand
data, such as having the contents box, quantity, ml and the same price.
 Mineral water data.

Image of Club brand mineral water data in each area with box, quantity,
and ml contents also have the same price data in every area of Jakarta,
Bekasi, depok and bogor.
 Chapter V

Implementation

In the picture above we explain using PCA clustergram k-means method.

Many clusters exist in our data, and there are 8 clusters of the data.
In the next picture we generate some plots to find a stable clusters
comparison and will be used for the determination of our data clusters, and
from the whole plot there are only 2 stable clusters.
For more specifications how many clusters are there in our data, the above
illustrates clusters 3 more stable in comparison with other clusters, and the
cluster maximum is in cluster 4.
From the picture above there are 7 clusters which show the plot stable
cluster 2, cluster 3, cluster 4, and cluster maximum 5.
 CHAPTER VI
Conclusion and Suggestions
6.1 Conclution

Scientificaly, data modeled using a NON-Hierarcy cluster then there is

obtained 8 Clusters as the number of cluster masimum. That gain is the
result of an iteration of the Non Hierarcy cluster algorithm. In the first
iteration, two clusters are generated and the next iteration forms a new
cluster. in the next stage applied a more in-depth analysis to determine the
level of stability of the cluster. From the test results found that cluster 3 is a
stable cluster so that the conclusion is in the RNA sequence data has 3
clusters of bacteria.

Based on The data that found two number of clusters. The whole plot
there are two stable clusters, its obtained that mineral water is a common
product. in the analysis shows that the products of each brand of aqua, vit,
lemineral, oasis, clif, sanqua have appeal to consumers. Most consumers
choose existing products that only look at prices, and there are also
consumers who see the product because of its quality and brand.
6.2 Suggestions
An objective this research is finding how many cluster that probably
in the data, and fixing number of stable cluster. The strength of this
research is dataset that collected isn’t just sample data but its actual data.
But for fixing problems not just a simple method. Kindly for further research
should be deep in analysis and fixing the visualisation to be a good
visualitation for data intepretation. A good visualization can make a model
easy to understand.
 Bibliography
(n.d.). Retrieved 07 29, 2018, from Proposal Eksperimen:
http://proposaleksperimen.blogspot.com/2011/12/proposal-ekperimen.html

(2018, 07 22). Retrieved from Open for Information Knime:

https://www.knime.com/knime-software/knime-analytics-platform

(2018, 07 22). Retrieved from R Studio:

https://www.rstudio.com/products/rstudio/

wikipedia.com. (2018).