Applying Call and Event Detail Records to Customer Segmentation and CLV

Applying Call and Event Detail Records to
Customer Segmentation and CLV
By:
Shohin Aheleroff
Shiraz University, Faculty of IT E-commerce

Abstract: Acquiring and retaining the most profitable
customers is a big concern of a telecommunication
operator to perform more targeted marketing therefore
business demand and competition between mobile
operators is becoming more based on life cycle of
customers in the network. In order to improve customer
satisfaction and fulfill requirements, several data mining
technologies can be used. Many researches have been
performed to calculate the customer value without
considering the call/event record generated according to
service usage while my research suggested analyses to
predict value of customer during their life based on CDRs
and EDRs that generated in the network. One of the most
important data mining technologies in life time value of
customers is customer segments. This targeting practice
has been proven effectively for mobile telecommunication
industry. Most operators evaluate their customers by
information like gender that extracted from billing
systems. This paper discusses an innovation to link
call/even detail record to the customer segmentation and
CLV for a telecommunication business. The subscribers
categorized in four segments (loyal groups) during the
CLV that influenced based on service usage.
I. INTRODUCTION
Mobile operator’s competitions and profits are facing
great challenges. The high volume of demand and
requirements associated with telecommunication services
has been changed and customers are demanding daily
basis. In order to improve response to customer
requirements, several data mining technologies can be
used. One of the most important data mining technics is
customer clustering analysis to categorize potential
customers into distinct groups for implementing
company’s roadmap and strategy. With the rapid growing
marketing business, data mining technology is indicate
more important role in the demands of analyzing and
utilizing the large scale information gathered from
enterprise data warehouse especially large amount call
detailed record of mobile subscribers. Dynamic
information about customer’s behavior is required to
segment and personalize products and services along with
business strategy and planning. This paper is an initiative
because almost always customer segmentation in
telecommunication business is only based on personal
(static) information such as age, gender and address from
a repository rather than from their actual (dynamic)

behavior. Furthermore, one of the key purposes of
customer segmentation and CLV is customer retention to
increase the loyalty and avoid churn volume. This paper
focused on proposing a customer segmentation and CLV
based on dynamic data from call/even data records.
II. Research Methodology
This study is designed to discover the application of
call and event detail records associated with service usage
by subscribers. I’m going to demonstrate that CLV and
customer segmentation will be best identified for a
telecommunication business by applying call/event detail
records than personal information extracted from a
repository such as a billing system. There are many
clustering method, for example, fuzzy clustering method,
system clustering method, dynamic clustering method to
be used however in order to evaluate the huge CDR/EDR
data, I take K-means clustering because it generates a
specific number of disjoint and flat (non-hierarchical)
clusters. The decision of how many customer segments a
company should create is largely dictated by the particular
make-up of their customers and the organizations ability
to develop and deliver unique segment specific marketing
treatments. In my research I will apply CDR/EDR data to
customer segmentation and CLV by using K-means as a
clustering tool. By considering the bellow steps, we need
enterprise hardware and software environment to deal
with huge amount of generated CDRs and EDRs but it’s
advised to select a sample of data to evaluate customer
habits and behaviors. Many segmentation algorithms
developed in software applications such as SAS and SPSS
but to make the model works the following steps advised
as the key outcome of my research:







Collecting CDRs & EDRs (by push or pull
mechanism into a data warehouse)
Selecting different services e.g. GPRS/MMS, voice ,
SMS and contents e.g. RBT, java Applications
Selecting key factors as the core items to monitor
customer‘s behavior.
Applying k-means as a well-known segmentation
algorithm.
Generating a matrix of segments as per each selected
factors.
Using the clustering output for loyalty and customer
churn application.

Due to huge volume of data generated by subscribers in
telecom and based on best marketing practices, generally
mobile operators are comfortable to have as few as five
unique segments, while other industries require as many
as twenty segments to satisfy their data-driven marketing
needs. The number of data in GSM is a barrier to analysis
customer‘s behavior, so almost there is a limitation to
analysis the whole data therefore a dynamic analysis for
limited duration based on detail records should be
considered and repeated continuously.





III. Prepare the data for clustering
I found a set of valuable information to identify core
needs in the life time of subscribers based on their call
detail record instead of their personal information such as
gender, address and income. A Call Detail/Data Record
contains at a minimum the following:











IV. Customer CLV definition:
According to the CDR/EDRs the interests, needs and
behavior of subscribers well-defined and distinct with
definition of each individual group of customer’s
behavior and related penetration in percentage:



Based on the level of loyalty of customers during their life
cycle (Fig.1) we really need to keep the plain loyal
motivated and also provide proper motivation and
package to improve their loyalty.

The number making the call (A number)
The number receiving the call (B number)
When the call started (date and time)
How long the call was (duration)
Call Type e.g. Voice call, SMS, GPRS/MMS
Balance before & after.
Location of mobile generator & terminator.
Incoming and outgoing Voice
Incoming and outgoing SMS
Different type of content

In addition to CDRs, I also considered subscribers who
are interested in activation or change any services by
capturing their record via analyzing EDRs (Event Detail
Records). By getting CDRs/EDRs from different sources,
we would be able to make sure that customers’ behavior
captured and we can evaluate what they are interested
more and less dynamically. Any call or event from
network elements such as IN and MSC will pass through
ODS via mediation , so by accessing to the repository of
xDRs and defining proper characters ,we build the model
of customer segmentation based on call/event detailed
record and their value on the network.



A customer has reached the Churn status for the first
time. He may in the future either stay in Churn status
or return to Active (he will then be labeled ‘Loyal
under Incentive’ from now on until he reaches again
and for good the Churn status).
Fence Seated:
A customer has moved out of Active situation into
Dormancy status for the first time. She/he may either
fall into the Churn status, remain Dormant, are
become Active again.
Loyal under Incentive:
A customer that has moved (once or several time) out
of Dormancy or Churn and back into Active status.

Plain Loyal:
A customer that has always been Active (never went
into Dormancy or Churn status).
Not Dependable:

Fig.1 Customer Life Cycle & Loyalty
In order to get a full picture of customer behavior in the
network and to realize their interest and also to predict
their needs, we will analysis the following two key
scenarios in detail based on historical call detail record:
o

o

Which segment he/she falls into and what are the
characteristics of this segment including CLV and
revenue value to the business.
What is the risk of this customer to leave the network or
remain inactive
V. Customers’ Behavioral Evolution
Based on six month historical data from enterprise
data warehouse, the statistics report of customers life
cycle shows (Fig.2) that more than 50 percent of plain
loyal subscribers are significantly decreasing while the
other there type of subscribers are not in the same level or
even increasing such as loyal under incentive subscribers.
I would like to highlight that due to behavior of
subscribers during the selected period of customers’ life
cycle, we would be in a position to predict that both loyal
plain and loyal under incentive subscribers will reach to a

single point. In this situation the two various segments
will merge to a unique group with population of 40
present of total subscribers. The volume of subscribers
(Fig.2) shows that the operator is quite a bit in a safe side
at this stage; however the monitoring of customers
behavior shows that we will face high rate of changes in
near future.

Fig.2 Customer Life Cycle
All the subscribers are active in the network till their
status will be changed to dormant and churned status. As
soon as they become a dormant subscriber, the risk will
be remained to leave the network and churn. The highest
challenge is related to the fence seated group because they
have potential capacity to change their status into loyal
incentive in a very optimistic view or they will leave the
network for ever (pessimistic).Besides the total number of
subscribers in each segment, we need to more focus on
the detailed information and track the dynamic behavior
of subscribers in each group and find a proper answer for
the following questions:
o
o
o
o

What happened to the ‘Fence Seated’ customers?
What happened to the ‘Not Dependable’ customers?
What happened to the ‘Loyal’ customers?
Where do each ‘’Loyalty type’ sit and what strategy
to apply?

Distinguish between groups, is the key milestone to make
distinguish promotion and motivation for each individual
groups. By the same way, marketing managements can
design more suitable marketing strategy. The behavior of
Fence Seated subscribers shows that all the existing
promotions and various tariff plan have not any impact on
this group, so as per detailed graph (Fig.3) after one
month only 41% of this subscribers remained in the same
situation while the 60% divided into two equal parts and
joint into “Not dependable” and “ Loyal under inventive”
groups. It’s so interesting that the after about four month
the loyal under incentive (66%) group members increased
to double compare to the Not dependable (33.5%) group
members.

Fig.3 Detailed Behavior of Customer Life Cycle & Loyalty
By focusing on not dependable subscribers which are
16% of total subscribers, it’s illustrated that only 10% of
these people moved to loyal under incentive while the rate
of movement increased up to 37% after four month. I
would like to highlight that not dependable staff only
moved to loyal under incentive group and not to any other
group. This is a good message to keep continue and boos
the existing marketing plan to increase the number of
loyal under incentive staff at the first step. The goal is to
make a plan to have specific motivation to move three

segments into the plain loyal group. The selected period
of subscribers’ life cycle is totally in a green status as
more than half out of total subscribers is plain total. There
is a big risk of churning to other operators because during
the last six month the trend of loyal subscribers is not
fluctuated and decreased to 49% that shows 8 % drop
down to other groups. Besides not dependable and fence
seated groups, the two other loyal groups have high rate
of upward (Loyal under Incentive from 19% to 33%) and
downward (Plain Loyal from 57% to 49%) changes.

the size of each loyal group , the average revenue per user
and the location of group in the matrix (Fig.4) , at least we
would be able to initiated a strategic plan because
sometime it’s very costly to motivate this group than put
more effort and cost on the loyal under incentive or fence
seated groups to make them plain loyal on the network, so
depend on the constraints such as budget , priority and the
number of subscribers in each group, management will be
able to make a decision accordingly.

VI. What strategy to apply?
Any operator required to build strong and profitable
customer relationships with solutions that increase
average revenue per user, reduce subscriber churn and
enhance brand loyalty. In order to utilize the best criteria,
two important parameters selected to identify level of
loyalty based on monthly recharge and active on net
(AON) for each segment of subscribers. The propagation
of subscribers and the location of each loyalty type will
lead the business to make an adequate plan, so based on
two mentioned items we mapped (Fig.4) four loyalty
groups into a matrix of duration (10 to 16 month) and
recharge (from 60$ to 100$) monthly basis. The margin
for AON is 13month and the recharge is 80$ monthly
basis. It means that if a subscriber is active more that 13
month on network, then its ether plain or under incentive
loyal subscribers depend on the volume of recharge per
month. By the same way if they are less that 13 month
active on network then they are either fence seated or Not
dependent, so it shows that we need development plan to
keep them more active on net by offering new packages
as motivations. As mentioned the boundary for monthly
revenue is 80$, so it’s quite important that even the
subscribers who charge more than specified range (80$)
they are not loyal. In addition to mentioned parameters to
identify behaviors and level of loyalty, we need to
consider the service usage and also the dependency or
relation between services as part of cross functional
management to improve customer segmentation. After we
recognized the location of loyal subscribers, it’s time to
plan the right strategy for each specific segment. The size
and the status of each loyal group have been illustrated in
the Fig.4 based on their monthly recharge and active time
on the network. Both fence seated and not dependable
subscribers have less size than other two groups and they
stayed around 12 month on the network as an active
subscriber, while fence seated staff pied more than
boundary. On the other hand, two plain and under
incentive loyal groups are almost 15 month active on the
net with different monthly recharge payment. It’s clearly
advised to motivate the loyal under incentive subscribers
to buy more vouchers (for prepaid) or bill payment
(postpaid) as the right strategy to make then plain loyal as
they are 33% of all subscribers. As a general concern and
risk, we might loss the entire 14% of not dependable
subscribers if they refuse to do payment. By considering

Fig.4 Loyalty segments based on their monthly recharge
VII. Delineate ‘Customer Life Value’
The CLV defined as the sum of the revenues gained from
company’s customers over the lifetime of transactions after
the deduction of the total cost of attracting, selling, and
servicing customers, taking into account the time value of
money .In order to identify the value that each type of
customers can bring into the picture, the value of mentioned
subscribers in one month calculated as following:

SEGMENT

Size

Individual
CLV

Estimated CLV
of the Base

Fence seated

5%

1,012

1,518,549,415

Plain Loyal

49%

1,343

19,755,580,218

Loyal under incentive

32%

1,134

11,228,482,296

Not dependable

14%

795

3,342,027,906

Table.1 Customer Life Value for each loyal group
(Monthly Revenue * No. months)
As identified earlier and shows in the Table.1 the statistic
report proved that the Loyal under Incentive customers
have 32% of total subscribers and the prediction of the net
profit is 1,134 which is less than Plain Loyal customers
with 49% population and 1,343 CLV as the highest net
profit segment of customers. The interesting highlighted

VIII. Extract Cross-Selling Value
The definition of Cross-selling and value generated in
the customer life time is selling new products to a
customer who has made other purchases earlier.in
telecommunication the cross-selling will reduce the churn
and keep subscribers as loyal as possible. Unlike the
acquiring of new subscribers, it involves either the
revenue increase from existing subscribers or it can
disturb the relation between service provider and the
customer. The validly and quality of customer
segmentation in a dynamic process that will help to
decrease the risk of negative impact of cross-selling and
associated value gained from individual segments.
In order to evaluate the service usage impact during
subscriber life time and its value, the majority of services
listed and a matrix illustrated in Table.2 to come up with
customer’s behavior associated with various services.









%

Voice

SMS

WOW

GPRS

RBT

F&F

Vitrin - TBS

Vitrin – Monotone

Vitrin - Polytonal

MMS

Roaming

point is that the higher CLV of a segment (Fence Seated)
compare to a bigger segment of customers which is 14%
clearly shows a group of subscribers with three time
smaller population generated 78% more as generated
1,012 net profit. As a conclusion we should not relay on
prediction and assumption because the CLV is dynamic
and might be differ due to the customers are using.
Similarly, the size of segment and the prediction of the
net profit of each individual don’t have direct relation to
anticipate and judge that the higher volume of customers
in each segment can generate the higher profit per
individual. Although the number of subscribers in each
segment will help to predict the profit but still there is no
guarantee to estimate until unless we get the information
regarding the different services and package they are
consuming in the network. This lead us to extract the
service usage information for different segment of
subscribers which is align with cross selling approach
across the network products and series.

Voice

100%

87%

42%

43%

16%

3%

12%

0.1%

0.03%

0.4%

0.1%

SMS

87%

100%%

39%

41%

16%

3%

12%

0.1%

0.03%

0.4%

0.1%

WOW

42%

39%

100%

21%

8%

2%

6%

0.1%

0.01%

0.2%

0.03%

GPRS

43%

41%

21%

100%

9%

2%

7%

0.1%

0.03%

0.3%

0.03%

RBT

16%

16%

8%

9%

100%

1%

4%

0.1%

0.01%

0.12%

0.01%

F&F

3%

3%

2%

2%

1%

100%

1%

0.01%

0.002% 0.04% 0.002%

Vitrin
TBS

12%

12%

6%

7%

4%

1%

100%

0.1%

0.01%

Vitrin
Mono

0.1%

0.1%

0.1%

0.1%

0.1%

0.01%

0.1%

100%

0.01%

0.1%

0.01%

0.003

Vitrin
0.03%
Pol

0.03%

0.01% 0.03% 0.01%

0.002%

0.01%

0.01%

MMS

0.4%

0.4%

0.2%

0.3%

0.1%

0.04%

0.1%

0.003% 0.002%

Roam

0.1%

0.1%

0.03%

0.03%

0.01%

0.002%

0.0001%

0.01% 0.0001% 0.00003% 0.0003

100%

0.002

0.00003%

100% 0.0003%

Table.2 Cross-Selling and services penetration
It’s also clearly reflected the high usage of a plan that
enables subscribers to make call or send SMS, MMS
more than the airtime amount of the recharge card
through WOW plan in a defined time period. I would like
to emphasis that an initiated campaign by considering
customer/subscriber life time will generate extensive
value even more than GPRS (internet) usage. Same result
generated by the mentioned Cross-Selling matrix for the
three type of content based services are the least services
that customers are willing to use after they joint into the
network.
IX. Categorize Revenue generated in CLV

Voice
SMS (Short Messaging Service)
GPRS ( General Packet Radio Service)
WOW ( service usage more than the airtime amount
of the recharge in a defined time period )
RBT (Ring Back Tone)
Content based services
MMS ( Multimedia Messaging Service)
Roaming

The matrix describes how two services related to each
other. For instance there is a likelihood of 87% for a
customer to use both voice and SMS, a likelihood of 43%
to use both voice and GPRS. Although the SMS usage is
not popular in most of the countries and operators, the
matrix shows that the subscribers are interested in short
message usage after making call.

We concentrated on the volume of services used by
subscribers and also their participation during customer
life cycle without considering the amount they pay via
different vouchers (air time) through various channels.
++
+
Avg
--

Greater than 50% ABOVE average
Between 25%-50% ABOVE average
within 25% of average
Between 25%-50% BELOW average
Greater than 50% BELOW average
Table.3 categorized values

In order to manage the situation and categorize the
volume / level of statistic reports, the thresholds provide
in Table.3 to have tangible report in five categories with
equal thresholds.

100%

87%

SMS

avg

avg

avg

Voice

avg

avg

avg

SMS

avg

avg
avg

avg

'Not Dependable'
(churn)

avg

'Fence seated'

avg

'Loyal Under
Incentive'
(dormancy)

avg

'Plain Loyal'

Voucher
Count

Voice

Services

Avg AON
(in months)

Voucher
Revenue

100%

SMS
Revenue

Services

Voice
Revenue

Penetration

REVENUE

avg

avg

avg

(-)

(-)

avg

(+)

++

--

0.1%

Roaming

+ +

avg

+ +

+

Roaming

avg

10%

TBS

+

avg

avg

+ +

TBS

(+)

++

(-)

--

--

0.1% Monotone

+

+

avg

+ +

Monotone

avg

(+)

(-)

(-)

(-)

0.02% Poly tone
12%
RBT

avg
+

avg
+

avg
+

+
+ +

Poly tone

avg

++

--

++

(-)

RBT

avg

(+)

(-)

(-)

(-)

0.4%

MMS

+ +

+ +

+ +

+ +

MMS

avg

++

--

(-)

--

45%

GPRS

avg

avg

avg

+

GPRS

avg

(+)

(-)

(-)

(-)

42%

WOW

+

+

+ +

+

WOW

avg

(+)

(-)

avg

(-)

3%

F&F
AVERAGE

+
88,637

+
4.7

+
56,368

+ +
16,722

F&F
AVERAGE

(+)

(+)

(-)

(-)

(-)

14.5

49%

33%

5%

14%

Table.4 Revenue generated based on voucher during CLV

Cluster7

avg
avg
+ +
‐
‐ ‐
‐
‐ ‐
+ +
‐
‐ ‐

avg
avg
‐
avg
avg
avg
avg
‐
avg
‐

avg
avg
( + )
( + )
+ +
( + )
+ +
+ +
( + )
+ +

avg
avg
+ +
‐
‐
‐
+ +
+ +
avg
avg

avg
‐ ‐
avg
‐ ‐
‐ ‐
‐ ‐
‐ ‐
‐ ‐
‐
‐ ‐

avg
avg
+ +
avg
( + )
+ +
avg
( + )
avg
avg

F&F

avg avg

AVERAGE 1%

‐

+ + ‐ ‐

Cluster1
0

Cluster6

avg
avg
+ +
avg
avg
( + )
( + )
avg
avg
avg

Cluster9

Cluster5

Voice
SMS
Roaming
TBS
Monotone
Poly tone
RBT
MMS
GPRS
WOW

Cluster8

SERVICE

Cluster4

The value of identified segments by considering service
usage during CLV has been exposed in Table.6
Cluster3

The more loyal subscriber results more revenue, therefore
it reduce churn in the life cycle of customers as per
Table.5.Except “Plain Loyal” segment of subscribers, the
rest have potential churn in the network. The TBS, Poly
Tone and MMS services consumed by loyal customers
while two of them (TBS & MMS) not even seen by nondependable customers. According to Table.5 the
customers who are active on net for long time are using
TBS and F&F more than other services. Moreover , it’s
amazing to find out that besides roaming the other
services consume below average by 33% of subscribers
that known as loyal under incentive, so they have enough
potential to leave the network because the service
threshold of this segment indicated below average service
usage during customer life cycle in this group.

XI. Identifying SEGMENTATION and CLV

Cluster2

X. Classifying Churn Propagation in CLV

According to the model presented in the Fing.9 the
visibility of churn based on level of participation and
service attraction is a sign that will be changed
dynamically and can’t be extended over CLV or to any
other telecommunication companies.

Cluster1

The amount of money that subscribers paid via voucher
(air time) for different services will generate useful
information that is hidden in the IN (intelligence network
e.g. CS5) data base. It’s so interesting that roaming
service with around 0.1 present and MMS with 0.4
percent usage in the network are part of most revenue
generated services. In addition it (Table.4) shows that the
most value extended from customers using Voice, SMS
and MMS during their life cycle in the network. The
number of vouchers consumed by MMS customers is
highest while the average is 4.5 vouchers for charging
prepaid accounts. Least but not last, the SMS consumers
who are using highest number of services (5 services) are
generating highest value and make revenue.

Table.5 Input from Churn during specific CLV

avg avg avg
‐ ‐ avg avg
‐ ‐ ‐ ‐ avg
‐ ‐ ‐ ‐ ( + )
‐ ‐ ‐ ‐ ( + )
‐ ‐ ‐ ‐ ( + )
‐ ‐ ‐ ‐ avg
‐ ‐ ‐ ‐ ( + )
‐ ‐ ‐ ‐ avg
‐ ‐ ‐ ‐ ( + )
‐ ‐ avg ‐ ‐ ‐ ‐ ( + )

2% 39% 17% 3% 8% 3% 3% 4% 21%

Table.6 Identify Customer Segmentation

XIV. REFERENCES
The highest number of customers is belong to cluster
No.3 with 39% and also average service usage in most
services and the next volume of customer are cluster 10
and cluster 4 with 21% and 17% which both are almost
equal to cluster 3 with 38% of total number of subscribers
in the network that makes revenue through consuming
various services. Besides multimedia and content based
services, voice and SMS are popular in all clusters within
25% of average usage. The behavior of customers…
Customer segmentation provides a model to categorize
the value generated by various groups of customers and
their level of participation to generate value from each
individual service.
II. CONCLUSION

[01] Ngai, E.W.T., Xiu, Li., Chau, D.C.K. (2008).
“Application of Data Mining Techniques in Customer
Relationship Management.” Expert Systems with
Applications, No.8:003-124.
[02] Kim, Su-Yeon. , Jung, Tae-Soo. , Suh, Eui-Ho., Hwang,
Hyun-Seok. (2006).“Customer segmentation and strategy
development based on customer lifetime value.” Expert
Systems with Applications, Vol.31:101–107.
[03] So Young, Sohn. Kim, Yoonseong. (2008). “Searching
customer patterns of mobile service using clustering and
quantitative association rule.” Expert Systems with
Applications, Vol.34:1070–1077.

Many researches attempt to improve customer
satisfaction by developing a comprehensive customer
segmentation model to gain highest profitability by analyzing
CLV. The mobile telecommunication marketplace is highly
competitive, so telecommunication companies often need to
design distinguishable marketing strategy based on different
behavior of their subscribers in order to improve their
marketing results and revenue. Call and Event Detail Records
is introduced as a key success factor to describe customer
behavior in this paper since it has valid and dynamic
information than a billing system. It is clearly demonstrated
that clustering analysis based on call and event details records
will give more enriched information than other analysis for
anticipating the business roadmap. This innovation proposed
how effectively can apply CDR/EDR data to customer
segmentation, customer life cycle; loyalty, churn and crossselling by considering the past contribution, potential value,
and churn probability at the same time. Three perspectives on
customer value (current value, potential value, and customer
loyalty) assist marketing managers in identifying customer’s
segmentation with more balanced viewpoints. After
identification of subscriber’s behavior and loyal groups, it’s
feasible and possible to put customer clusters in place and
make an applicable strategic plan for each group to achieve
highest customer satisfaction along with maximum revenue
for telecommunication industries.

[04] McCarty, John A., Hastak, Manoj. (2008). “Segmentation
approaches in data-mining: A comparison of RFM,
CHAID, and logistic regression.“Journal of Business
Research, Vol.60, No. 6:656-662.

XIII. ACKNOWLEDGMENT

[10] Charles Dennis, David Marsland, Tony Cockett. (2001).
“Data mining for shopping centers - Customer knowledge
management framework.” Journal of Knowledge
Management, Vol.5, No.4:368-374.

The author would like to thank Dr. Gholamian for his support,
supervision and direction in University of Science &
Technology - IUST. My supervisor always tracks the work to
make sure there is no problem, and if there is, he would give
immediate help to solve the issue regarding my research.
Special thanks to my colleagues for the great encouragement
and help on both studies and work. Thanks to my wife,
Somayeh always lets me know she loves me which gives me
motivation. At last, I want to thank International Journal of
Information and Communication Technology for the
opportunity to share the knowledge.

[05] Hung, Shin-Yuan. , Yen, David C., Wang, Hsiu-Yu.
(2008). “Applying data mining to telecom churn
management.”Expert Systems with Applications,
Vol.31:515–524.
[06] Han, J., Kamber, M. (2006), “Data mining: Concepts and
Techniques”, USA, Morgan Kaufmann Publishers.
[07] Leea, Jang Hee., Park, Sang Chan. (2005).”Intelligent
profitable customer segmentation system based on
business intelligence tools.” Expert Systems with
Applications, Vol.29:145–152.
[08] Jain, D., & Singh, S. S. (2002).”Customer lifetime value
research in marketing: a review and future
directions.”Journal of Interactive
Marketing, Vol.16, No.2, 34–45.
[09] Mali, K. (2003).“Clustering and its validation in a
symbolic framework.” Expert Systems with Applications,
Vol. 24:2367-2376.

[11] Bayo´n, T., Gutsche, J., & Bauer, H. (2002).”Customer
equity
marketing:
touching
the
intangible”
EuropeanManagement Journal, 20(3), 213–222.
[12] Shohin Aheleroff, (2010). “Customer Segmentation for a
Mobile Telecommunications Company Based on Service
Usage, International Journal of Convergence Information
Technology 308 – 313, ISBN: 978-1-4673-0231-9

Applying Call and Event Detail Records to Customer Segmentation and CLV

More Related Content

What's hot (16)

Viewers also liked (14)

Similar to Applying Call and Event Detail Records to Customer Segmentation and CLV (20)

Recently uploaded (20)

Applying Call and Event Detail Records to Customer Segmentation and CLV