Using Text or Short Text Mining for Optimizing Customer Relationship Management

This paper discusses the use of text mining or short text mining for optimizing customer relationship management. It explores the importance of text mining in business intelligence and the different methods of text mining. The survey assesses how text mining can improve customer relationship management and the technical details of the process.

Running head: TEXT OR SHORT TEXT MINING
Using Text or Short Text Mining for Optimizing Customer Relationship Management
Name of the Student
Name of the University
Author Note

2TEXT OR SHORT TEXT MINING
Abstract:
The main aim of this paper is conducting a survey on the text mining or the short text mining in
the aspect of customer relationship management, whether it is capable of optimizing customer
relationship management or not. First a brief discussion has been done in this paper regarding the
short text mining process. In this section the short text mining has been understood briefly. In the
survey part of this paper first the importance of this survey has been elaborated. In the next
section of this survey how the text mining is capable of improving the customer relationship
management has been discussed. Important steps for optimizing the customer relationship
management has been discussed here. This paper also has discussion regarding the technical
details of the text mining process. It has been also assessed that there are various of ways in
which text mining can be utilized in the business intelligence. There are mainly four working
procedures of the text mining that has been discussed in this report and it has been assessed that
pattern taxonomy method is the most effective among the four.

3TEXT OR SHORT TEXT MINING
Table of Contents
Introduction:....................................................................................................................................4
Survey on the Text or Short Text Mining:.......................................................................................5
Importance of this Survey:...........................................................................................................5
How Text Mining optimizes Customer Relationship Management:...........................................6
Text Mining in Business Intelligence:.........................................................................................9
Differences among Methods:.....................................................................................................10
Conclusion:....................................................................................................................................12
References:....................................................................................................................................14

4TEXT OR SHORT TEXT MINING
Introduction:
The text mining which is again refereed as the short text mining is actually relevant with
the text analytics. The text mining is the process of high quality information derivation from
some specific texts (Weiss, Indurkhya and Zhang 2015). Using the text mining the high quality
data is achieved through pattern and trends devising. One of the prior example in this case is the
statistical learnings of patterns. The process of text mining mainly involves with the process of
input text structuring, patter deriving among the structured data and evaluation and interpretation
of the data that has been extracted (Allahyari et al. 2017). Normal process of the text mining
includes clustering of the texts, categorization of the texts, extraction of entity, granular
taxonomies production, summarization of the documents and sentiment analysis. The word text
analytics demonstrates a set of linguistic, statistical, and the techniques of machine learning
which structure and model content of information of textual sources of exploratory data analysis,
business intelligence and investigation. The text mining also describes text analytics application
so that response can be provided against the business problem (Isayev 2019). Response can be
provided in both independent way or in conjugation with analysis and query fielded numerical
type of data. The techniques of text mining is capable of discovering and presenting knowledge,
facts, relationships and business rules which is locked in textual formats.
The method of text mining plays an crucial role in the context of improving the customer
relationship management which is quite important for the business. Through the improvements
in the customer relationship management the overall business procedures for an organization can
be improved easily (Soltani and Navimipour 2016). The customer relationship management is
actually a strategy of management for all the relationship that the organization currently have.
Here the important relationship which an organization possess is with its customers and

5TEXT OR SHORT TEXT MINING
stakeholders. The customer relationship management is important for improving the profitability
of the organization. Proper customer relationship management helps to focus on the relationship
procedures of the organization with the individuals (Rahimi and Kozak 2017). These individuals
can be customers, suppliers and service users. The customer relationship management or the
CRM is important for achieving some biggest gains in the case of productivity of the
organization.
The CRM and the text mining has a deep relationship among them. The customer
relationship management is very much important for the businesses while the text mining is
capable of improving the CRM. Thus, in other way the text mining or the short text mining plays
an important role for improving the overall business procedures. Thus in this paper a survey will
be done regarding how the process of text mining can optimize the customer relationship
management for the businesses (Navimipour and Soltani 2016). In this survey how the text
mining is capable of optimizing the customer relationship will be demonstrated. In the following
section how this can be utilized in the business intelligence will be discussed thoroughly. Also,
differences among several methods for optimizing the customer relationship management will be
discussed in this context.
Survey on the Text or Short Text Mining:
In this report a survey will be done on the process of text mining or the short text mining
regarding how it can optimize the process of customer relationship management for the
businesses. In the following section of this report, all the important context of this report will be
demonstrated.

6TEXT OR SHORT TEXT MINING
Importance of this Survey:
This survey is all about the technique of enhancing the procedures of the customer
relationship management using the text or short text mining. Through this survey this paper
actually aims to find out whether the text mining is really capable of improving the business
intelligence or not. With that, comparison will be also done with the other process of optimizing
the customer relationship management. Through this survey importance of the text mining in the
aspect of businesses can be identified. Due to this factor this survey on the text or the short text
mining is very much important in this context.
How Text Mining optimizes Customer Relationship Management:
The working procedure of the text mining is very much unique in nature. Due to this
unique factor, the text mining is capable of optimizing the customer relationship management.
The process of text mining is useful for the analysis of textual content in the real time (Silge and
Robinson 2017). This principle of text mining is actually used for the optimization of the
customer relationship management. Customers all over the world tends to provide feedbacks
which mainly comes in the textual formats. This textual formats can be analyzed using the text
mining technique so that analysis of those textual contents can be done on a real time basis
(Vijayarani, Ilamathi and Nithya 2015). In this case through the analysis of those textual contents
sentiment trends of the customers can be identified and from that the organizations will be able
to adjust their message of marketing. In this way the method of text mining is capable of
optimizing the customer relationship management heavily. With that, the monitoring of
sentiments of the customers is also important in this context as this will highlight the major
issues with the products or the services and from there important feedback from the customers
can be collected. Collecting the feedback only will not solve the problem in this context. These

7TEXT OR SHORT TEXT MINING
feedbacks needs to analyze properly and in most of the cases these feedbacks comes in a large
quantity. For analysis of this large amount of data text mining is the very much useful and for
this reason actually the customer relationship management will be optimized (Zhou et al. 2016).
Thus, in this way the text mining optimizes the customer relationship management.
Now the important question is how the text mining analyses those large quantity of data.
There are mainly four ways through which the process of text mining works. These four methods
of text mining are the phrase based method, term based method, pattern taxonomy method and
concept based method. Though there are different methods for the text mining, the working
procedure for the all the text mining are the same. From the above discussion it has been already
assessed that the text mining helps to understand the texts in much more better way. This process
exchanges the words from unstructured type of data to numerical type of values (Fleuren and
Alkema 2015). Text mining works in this aspect by identification of the relationships and
patterns which relies within a huge amount of data. In most of the cases the text mining uses
computational algorithm for reading and analysis of text information. Without implementation of
the text mining it will be become quite problematic to understand the texts quickly and easily
(Krallinger et al. 2017). It is possible to mine the texts in a more comprehensive and systematic
way and in this aspect all the information regarding the business can be gathered automatically.
The important steps for the text mining process are the information retrieval, natural language
processing, information extraction and data mining.
The information retrieval is the first step for the text mining for optimization of the
customer relationship management (Debortoli et al. 2016). In the step of information retrieval a
search engine is utilized for finding the collection of texts which are mainly known as the corpus
of texts. In this aspect the corpus of texts might need some specific type of conversion. In this

8TEXT OR SHORT TEXT MINING
step all these texts needs to be brought together in a specific type of format that will be very
much useful for the users for the understanding purposes. In most of the cases the XML format is
utilized as a standard of text mining.
One of the important step of text mining is the natural language processing. This step of
text mining is important as this allows the whole system to conduct a grammatical analysis of the
sentences for reading the texts. In this step analysis of the texts is also done in structures.
After the natural language processing is done the second step of the text mining occurs
which is information extraction. This stage of text mining is important as in this particular step
meaning of a specific text markup is identified (Alghamdi and Alfalqi 2015). In this step, also
the process of adding locations and names to the texts is done. In this particular step search
engine is able to extract the information and is able to find the relationships among the texts and
for that metadata of those texts are utilized.
The third and the final step for the text mining is the data mining for which different kind
tools are required. In this step similarities among the information is assessed. Here similarities
means same type of meaning which are very much difficult to find (Kumar and Ravi 2016). Text
mining is a tool which is capable of boosting the process of research and it also assists in testing
of the queries.
The text mining has some important elements which are very much important for
optimization of the customer relationship management. These includes granular taxonomies,
document summarization, text clustering, text categorization, sentiment analysis and entity
relationship modelling.

9TEXT OR SHORT TEXT MINING
Text Mining in Business Intelligence:
Text mining plays an crucial role in the aspects of business intelligence. Through the
implementation of text mining in the business intelligence organization can achieve competitive
advantages (Laursen and Thorlund 2016). The text mining is actually a process of automated
extraction of important information form a huge unstructured texts. In this process of text mining
identification of the patterns are done from a natural language text instead of any traditional
databases in aspect of data mining. The text mining is also different from the process of web
mining as text mining is normally performed over the unstructured type of data so that genuine
information can be extracted from those unstructured type of data. Unknown and crucial data is
discovered using the process of data mining for the future predictions and this is very much
important for the businesses.
The word text mining is used for denoting the systems which is capable of analyzing
huge amount of natural language text by parsing the whole text. In this process following the
parsing of the information lexical usage patterns are utilized for the extraction of correct
information (Ngai and Lee 2016). The text mining searches for the patterns within the texts by
the automatic extraction. Linking of the gathered information is very much important as this
helps to form some new hypothesis and facts which can be explored in the future by performing
experiments. The process of text mining is quite different when it is compared with process of
data mining as data mining indicates extraction of previously unknown, implicit and useful
information from the data where the text mining the data which needs to be extracted is not
hidden at all and not explicit in nature (Tan 2018). The text mining explores a typical type of
data which can be directly pushed into the data without any type of intervention of the humans.
The technique of text mining has evolved from some related technologies which are based on

10TEXT OR SHORT TEXT MINING
probability theory, statistics and artificial intelligence. For an example performance of any
random searches can be identified using counting failures and the success.
Text mining successfully displays flexible approaches of analysis, research and
information management on the textual type of materials. The process of text mining also assists
in processing of the natural languages as it is capable of automating the analysis of links
regarding citations within the texts and hyperlinks within the webs. The natural language is only
dependent on common sense knowledge which is very much difficult to encode in an algorithmic
format (Young et al. 2018). So, text mining comes with outgrowth of real-text mindset that
provides shallow processes of unrestricted type of texts and a deep type of processing of the
materials that are specific to the domain.
Thus, in this aspects the text mining can be used in the aspect of business intelligence.
The important applications of the text mining within the business intelligence are the scientific
analysis of data, biomedical sciences, fraud detection and document warehouse for SAP (Larson
and Chang 2016). With that the text mining can be also utilized for marketing purposes in the
aspects of business intelligence. With that market segmentation can be done and potential
customers can be identified using the text mining in the aspect of business intelligence which
will also optimize the customer relationship management.
Differences among Methods:
The text or the short text mining has a huge impact in improving the customer
relationship management. For optimizing the customer relationship management the text or short
text mining works in four different ways which are the term based method, phrase based method,
concept based method and pattern taxonomy methods. In the following section each of these
methods are described briefly.

11TEXT OR SHORT TEXT MINING
 Term Based Method: Terms in the documents always have a semantic type of meaning.
In the term based method analysis of the whole document is done on the basis of terms.
Due to this factor there is one advantage in this method which is efficient computational
performance (Kobayashi et al. 2018). With that another advantage in this case is mature
theories for term weighting. Though there are several of advantages, the main
disadvantage is problems regarding polysemy and synonymy. Polysemy occurs when
single words has multiple of meaning and synonymy occurs when multiple words are
having same meaning. Phrase Based Method: Phrases consists more of the semantics like the information and
less ambiguous in nature. This method of text mining works based on phrases which have
more meaning and less obvious and more discriminative in nature (Bholat et al. 2015).
This method of text mining is having several of disadvantages which are having low
frequency of occurrence, inferior statistical properties to terms and having large number
of noisy phrases. Concept Based Method: In this method, analysis of the document is done depending on
the document and sentence level. First component is responsible for examining the
meaningful parts of the sentences, while the second component is responsible for
producing a conceptual ontological graph so that structures can be explained (Salloum et
al. 2017). The third component is responsible for extracting the top concepts that are
dependent on first two components. The main advantage of this method is that it is
capable of differentiating between the words such that which are the important one which
are the unimportant one.

12TEXT OR SHORT TEXT MINING
 Pattern Taxonomy Method: In this method of text mining analysis of the documents are
done depending on the patterns. Patters in the document can be discovered by utilization
of techniques of data matching which includes sequential pattern mining, association rule
mining, closed pattern mining and frequent item set mining. Two processes are used in
this method of text mining which are pattern evolving and pattern deploying process.
This method of data mining faces problems regarding low frequency (Irfan et al. 2015).
Also, misinterpretation of the patters occurs in this type of method. Though this method
of data mining is having some problems still this is much better method of text mining
when compared with the other three methods of text mining.
Conclusion:
From the above discussion it can be concluded that text mining or the short text mining is
one of the crucial aspects for optimizing the customer relationship management which will
actually improve the performance of the organization. From the report it has been assessed that
text mining is actually a specific way of information derivation form a large quantity of source.
In this paper a survey has been done regarding the text mining or short text mining. In the
introductory part of this assignment a brief description of the text mining has been demonstrated
and how the text mining is related with customer relationship management has been discussed.
In the following section of this report survey on the short text mining has been done. In this
survey part first of all why this survey is important has been elaborated. In the following section
of this report the technical aspects of the text mining has been discussed. In the technical aspects
how this system works has been discussed. In the further section of this report, how the text
mining can be utilized in the aspects of business intelligence has been described briefly. In this
section it has been analyzed that there are various of aspects due to which text mining is

13TEXT OR SHORT TEXT MINING
important in the context of business intelligence. In this aspects it can provide competitive
advantage to the organizations. In the last section of this survey differences among various
working methods of text mining has been discussed. It has been assessed there were four
different kind of methods which were term based method, phrase based method, concept based
method and pattern taxonomy methods. All of those methods has been analyzed briefly and it has
been assessed that pattern taxonomy method is the most efficient method among all of the four
methods.

14TEXT OR SHORT TEXT MINING
References:
Alghamdi, R. and Alfalqi, K., 2015. A survey of topic modeling in text mining. Int. J. Adv.
Comput. Sci. Appl.(IJACSA), 6(1).
Allahyari, M., Pouriyeh, S., Assefi, M., Safaei, S., Trippe, E.D., Gutierrez, J.B. and Kochut, K.,
2017. A brief survey of text mining: Classification, clustering and extraction techniques. arXiv
preprint arXiv:1707.02919.
Bholat, D., Hansen, S., Santos, P. and Schonhardt-Bailey, C., 2015. Text mining for central
banks. Available at SSRN 2624811.
Debortoli, S., Müller, O., Junglas, I. and vom Brocke, J., 2016. Text mining for information
systems researchers: An annotated topic modeling tutorial. Communications of the Association
for Information Systems, 39(1), p.7.
Fleuren, W.W. and Alkema, W., 2015. Application of text mining in the biomedical
domain. Methods, 74, pp.97-106.
Irfan, R., King, C.K., Grages, D., Ewen, S., Khan, S.U., Madani, S.A., Kolodziej, J., Wang, L.,
Chen, D., Rayes, A. and Tziritas, N., 2015. A survey on text mining in social networks. The
Knowledge Engineering Review, 30(2), pp.157-170.
Isayev, O., 2019. Text mining facilitates materials discovery.
Kobayashi, V.B., Mol, S.T., Berkers, H.A., Kismihók, G. and Den Hartog, D.N., 2018. Text
mining in organizational research. Organizational research methods, 21(3), pp.733-765.
Krallinger, M., Rabal, O., Lourenco, A., Oyarzabal, J. and Valencia, A., 2017. Information
retrieval and text mining technologies for chemistry. Chemical reviews, 117(12), pp.7673-7761.

15TEXT OR SHORT TEXT MINING
Kumar, B.S. and Ravi, V., 2016. A survey of the applications of text mining in financial
domain. Knowledge-Based Systems, 114, pp.128-147.
Larson, D. and Chang, V., 2016. A review and future direction of agile, business intelligence,
analytics and data science. International Journal of Information Management, 36(5), pp.700-
710.
Laursen, G.H. and Thorlund, J., 2016. Business analytics for managers: Taking business
intelligence beyond reporting. John Wiley & Sons.
Navimipour, N.J. and Soltani, Z., 2016. The impact of cost, technology acceptance and
employees' satisfaction on the effectiveness of the electronic customer relationship management
systems. Computers in Human Behavior, 55, pp.1052-1066.
Ngai, E.W. and Lee, P.T.Y., 2016, June. A Review of the literature on Applications of Text
Mining in Policy Making. In PACIS (p. 343).
Rahimi, R. and Kozak, M., 2017. Impact of customer relationship management on customer
satisfaction: The case of a budget hotel chain. Journal of Travel & Tourism Marketing, 34(1),
pp.40-51.
Salloum, S.A., Al-Emran, M., Monem, A.A. and Shaalan, K., 2017. A survey of text mining in
social media: facebook and twitter perspectives. Adv. Sci. Technol. Eng. Syst. J, 2(1), pp.127-
133.
Silge, J. and Robinson, D., 2017. Text mining with R: A tidy approach. " O'Reilly Media, Inc.".

16TEXT OR SHORT TEXT MINING
Soltani, Z. and Navimipour, N.J., 2016. Customer relationship management mechanisms: A
systematic review of the state of the art literature and recommendations for future
research. Computers in Human Behavior, 61, pp.667-688.
Tan, P.N., 2018. Introduction to data mining. Pearson Education India.
Vijayarani, S., Ilamathi, M.J. and Nithya, M., 2015. Preprocessing techniques for text mining-an
overview. International Journal of Computer Science & Communication Networks, 5(1), pp.7-
16.
Weiss, S.M., Indurkhya, N. and Zhang, T., 2015. Fundamentals of predictive text mining.
Springer.
Young, T., Hazarika, D., Poria, S. and Cambria, E., 2018. Recent trends in deep learning based
natural language processing. ieee Computational intelligenCe magazine, 13(3), pp.55-75.
Zhou, Y., Tong, Y., Gu, R. and Gall, H., 2016. Combining text mining and data mining for bug
report classification. Journal of Software: Evolution and Process, 28(3), pp.150-176.