<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Impact of Data Mining and Social Media Marketing to Enrich Customer Satisfaction</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Mohamed A. Hamada</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Adejor E. Abiche</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>International Information Technology University</institution>
          ,
          <addr-line>Manas St. 34/1, Almaty, 050040</addr-line>
          ,
          <country country="KZ">Kazakhstan</country>
        </aff>
      </contrib-group>
      <abstract>
        <p>Data mining techniques are intensively utilized to process all the collected data quickly and support firms in being competitive in different markets. Customer satisfaction rate is highly dependent on the ability of businesses to manipulate huge amounts of information and make appropriate decisions based on such manipulation. Supplying a unique algorithm for targeted advertisement and account promotion based on Big Data analysis has allowed Instagram to become a company that successfully meets the demand of both its B2B and end-consumers. This research aims to demonstrate how social media uses data and data mining to bring value to the customers and increase their satisfaction level on the example of Instagram. Quantitative and qualitative study is conducted to support research with primary data input. The sample size for quantitative analysis is 106 Instagram users, aged 16+, living in Kazakhstan. Google form questionnaires were used to collect answers. During the research, several study development and opportunity areas were revealed to continue the in-depth investigation of the case. One such question is data privacy which can also be studied on the example of several social networks.</p>
      </abstract>
      <kwd-group>
        <kwd>1 Social Media</kwd>
        <kwd>Instagram</kwd>
        <kwd>target advertisement</kwd>
        <kwd>data mining</kwd>
        <kwd>customer satisfaction</kwd>
        <kwd>social media marketing (SMM)</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>their strategy from offline to online. Social media platforms allow companies to build their brand
image, receive customer feedback, organize marketing campaigns and social interaction with
their buyers. Based on these data, businesses monitor such metrics as customer and post
engagement rates, impressions and reach, click-through rate (CTR) in ads and posts, lead
conversion rate, and response rate, and receive a 360-degree view of online performance based
on interaction with other social media users. These data can be classified as received as the result
of user activity.</p>
      <p>At the same time, each user gives applications and websites consent for certain data collection
and processing. Such permission assumes collecting information on age, gender, preferences,
geolocation, social interaction, list of purchases, things you search for, ads you click, websites you
visit and even devices you use to access the site. Furthermore, some applications use the
microphone and location history. Gathering such data, companies use data mining to get a
broader understanding of customer profiles. The result of thorough customer data analysis is an
example of successful marketing campaigns and effective social media interactions and presence.
Such examples include two international companies - e-commerce giant Amazon and
multinational beauty retailer Sephora, and two local companies - fintech firm, Kaspi, and internet
corporation Yandex.</p>
      <p>One of the first companies to effectively utilize social media and data mining in combination is
Amazon.com. The company is in the range of first movers to build interactive and diverse social
media strategies. Amazon's website platform is user-friendly, easy to navigate and sensitive to
consumer behaviour. Amazon is one of the pioneers, which started to introduce service/goods
recommendations based on users' previous orders, clicks and overall behaviour. The company
became a leader in the online retail sector because it was successful in understanding the
opportunity of e-commerce and personalized approach and could implement it using online
platforms. Currently, Amazon.com uses data mining techniques and continuously refines its social
media marketing to meet complicated customer demands and tastes.</p>
      <p>In addition, Sephora is an international retailer, which accurately develops its social media
marketing, which has brought more of its brands onto Instagram checkout to make clients’
shopping faster, easier, and more reliably convenient. It has a wide social media presence having
a website, mobile app, and social networks. The main feature of its social media marketing (SMM)
strategy is that all the components are smoothly operated together which leads the company to
one goal. Successful social media presence allowed Sephora to build a strong brand community
which contributes to the higher clients' retention and satisfaction rates.</p>
      <p>The most demonstrative examples of prosperous companies in the Kazakhstan market, that
use data mining and social media to deliver value and increase customer satisfaction level are
Kaspi. kz and Yandex. kz. Kaspi. kz uses several sources which collect data about consumer
behaviour, such as websites, mobile applications and social networks. Kaspi mobile applications
for end-users and B2B are popular among the population because they are convenient to use and
personalized. Kaspi was the first bank in Kazakhstan to introduce the application, which
combines many services for the consumer, such as money transfers, payment transferring, online
shopping, and online deposit/credit opening. Having all these services inside the mobile phone
helps to solve several issues and make the clients' lives faster and easier. Kaspi at the same time
gained the tool for reliable data collection, which further is analyzed by data mining techniques
to visualize the data and use it for decision making. The Kaspi consumers are satisfied and the
number of users is increasing, meaning that the company is now on the right way, but it should
always be on track to save its superiority. The recent step to meet customer demand from the
Kaspi side was to introduce mobile POS for B2B to accept payments from end-consumers. As a
result, the bank is gaining a bigger pie of market share by introducing more convenient products
to customers and expanding its social media presence.</p>
      <p>Yandex is another example of a company that tries many different fields of digital business.
Yandex is a competitor of Google in the post-soviet market. It has such services as a search engine,
cloud data storage, taxi mobile app, food delivery service, carts etc. On the local Kazakhstani
market, Yandex is well-known for its mobile app, which combines several services in one - taxi
and food delivery. Yandex is using geo data to determine the location of an individual. It uses
previously obtained data about consumer behaviour to recommend products and services for
future requests. Yandex is always testing new products, sometimes innovations become
successful, and some of them fail. However, the company itself is advanced in terms of using Big
Data and data mining to support its business decisions.</p>
      <p>
        The companies described above have their accounts on various social networks. It is suggested
that 72% of people will access the global network using only smartphones by 2025 [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ].
Companies need to be present in social networks to effectively communicate with their
consumers. Almost all thriving companies have pages on Instagram, as it is one of the most
popular social networks. Instagram opens opportunities not only to its end-consumers but also
to B2B clients. End-consumers can share photos, and videos, purchase necessary services/goods.
B2B clients receive the marketplace to sell the products and get the chance to have all the needed
statistics about their consumers. Instagram developed its business model, which combined
entertainment and commercial concepts in one place satisfying many types of customers at one
time. Kazakhstan is in first place on Instagram reach worldwide - 72% of all the internet users in
Kazakhstan are Instagram users [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ].
      </p>
      <p>This paper aims to analyze how Instagram uses data collected from its users to increase their
satisfaction with the product.</p>
    </sec>
    <sec id="sec-2">
      <title>2. Literature review</title>
      <p>
        The major challenge for a business is the accurate planning of new product launching and
forecasting its further lifecycle. Before product launching, a company should find out the key
features of targeted audiences. These findings will help to set up appropriate pricing politics and
forecast sales. The next step is to develop forecasting actions based on prediction analysis to
ensure that the product will stay relative to the target audience. Modern technologies such as
Instagram assist with this challenge, solving it most efficiently and effectively [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ].
      </p>
      <p>
        Business analytics is a process that is essential for modern business development. The point
of business analytics is organizing quantitative data and transforming it into information for
decision-making purposes. Business analytics assumes converting data patterns into
visualization and reporting formats. Implementing this practice gives managers a clear view of
what happened in the company with the reporting date and what is going to happen in the future.
Reports can be created and visualized from any perspective, including financial, operational etc.
[
        <xref ref-type="bibr" rid="ref5">5</xref>
        ].
      </p>
      <p>
        One of the advanced tools of business analytics is data mining. Data mining can be used for any
field of study; however, this research covers the effectiveness of data mining in obtaining and
measuring client satisfaction. Data mining is a powerful technique for understanding customer
satisfaction patterns and predicting future customer behaviour in various business fields. For
example, using data mining tools such as decision trees and neural networks for the fast-food
sector gives accurate results for identifying key customer satisfaction criteria with 80%
correctness [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ] Tama. At the same time, the banking system is one of the most dynamic and
competitive big data user industries, experiencing high demand for data mining techniques. As a
result, banks are prosperous in utilizing data mining applications, they usually use more than two
mining tools to analyze vast amounts of big banking data [
        <xref ref-type="bibr" rid="ref7">7</xref>
        ]. In the retail business field, building
a marketing strategy with a focus on customer experience and preferences is key to increased
sales and better company performance. Data mining facilitates client-oriented marketing in retail
companies, which helps not only attract but also retain consumers [
        <xref ref-type="bibr" rid="ref8">8</xref>
        ]. Finally, businesses reduce
costs and increase customer satisfaction and brand equity. Based on the information above it can
be concluded that data mining techniques are vital for many industries. Furthermore, companies
from diverse business sectors are willing and able to implement these tools to build a
comfortable, delightful and interactive consumer journey.
      </p>
      <p>
        As social media plays a significant role in building general marketing strategy, the most
successful companies are actively engaged in creating effective social media marketing (SMM)
strategies. It has proven that properly implemented SMM has a drastic impact on perceived brand
value and as a result leads to higher client satisfaction levels [
        <xref ref-type="bibr" rid="ref8">8</xref>
        ]. Social media analytics becomes
a powerful tool in establishing the main marketing framework based on previous consumer
behaviour. Such kind of analytical data is collected from the first source and gives a reliable
foundation for further analysis. E-business particularly e-commerce as recently developed
business models, contribute to the fast collection of huge amounts of initial data. The challenge is
that analyzing high volumes of data without appropriate tools becomes difficult [
        <xref ref-type="bibr" rid="ref9">9</xref>
        ].
      </p>
      <p>
        Data collected from social media is used for customer analysis. Customer analytics is a key
process for making crucial business decisions through marketing segmentation and predictive
analysis. The collected information includes such data as customer lifestyle, demographics,
preference values, location information, goods purchased, and payment methods. It is a major
benefit for companies in respect of monitoring market fluctuations and trends. Innovative
marketing strategies are developed based on customer analysis, giving a company a competitive
advantage over competitors. From the client's perspective, customer profiling assists with
achieving common values and expectations between the company and the customer, and in turn,
increases the accuracy of customer satisfaction [
        <xref ref-type="bibr" rid="ref10">10</xref>
        ].
      </p>
      <p>
        As a result of customer analysis, the company establishes customer profiling, which defines a
description of the ideal customer for its business. Using this tool, a company understands its
customers better and targets them in the most effective way [
        <xref ref-type="bibr" rid="ref11">11</xref>
        ].
      </p>
      <p>
        One example of social media data mining is text mining which allows the identification of the
specific topics and themes discussed in the text based on keywords and terms. In the text mining
techniques review of two huge social networks Facebook and Twitter, it was revealed that data
in the English language is easier to be interpreted and analyzed [
        <xref ref-type="bibr" rid="ref12">12</xref>
        ]. Based on the analysis it can
be concluded that the success of the decoding process highly depends on the language. In general,
data provided for data mining should be gathered in a universal format.
      </p>
      <p>
        Based on customer profiles companies launch target advertisements for their products and
services. Data mining with cluster analysis indicates two clusters - price-caring and
productcaring - based on customer's reactions and buying behaviour once they see the advertisement
[
        <xref ref-type="bibr" rid="ref13">13</xref>
        ].
      </p>
      <p>
        Company profile/image in social media has a direct impact on the level of customer
satisfaction as brands are no more abstract ideas, they have people behind them and a story that
is broadcasted via their internet pages. Research showed that the highest correlation between
social media analytics and customer satisfaction is found in the case of a company operating in a
highly competitive local environment [
        <xref ref-type="bibr" rid="ref14">14</xref>
        ]. This is conditioned by the huge amount of local
unknown companies while building brand image in social media allows one of them to stand out
and build strong relationships with customers and create a positive brand image.
      </p>
      <p>
        Big data gathered from social media goes further than just a tool used for determining the
advertising content, marketing campaigns or customer profiles. It does not end with promotions
and interactive content based on views and likes. Social networks can give enough information to
governmental organizations and international corporations. Researchers and statistical agencies
also mine profiles on different social networks. Based on such data agencies can market political
ideologies, sociocultural changes, and religious groups [
        <xref ref-type="bibr" rid="ref15">15</xref>
        ]. Based on the location and migration
of people during natural calamities rescue groups managed to find the victims and save them.
      </p>
      <p>
        The effective methods of using social media and data mining were identified and evaluated in
the educational sector to improve the level of customer satisfaction through analyzing reports
and collecting data about opinions on adopting data mining and social media [
        <xref ref-type="bibr" rid="ref16">16</xref>
        ].
      </p>
    </sec>
    <sec id="sec-3">
      <title>3. Research methodologies</title>
      <p>The relationship between data mining, social media and customer satisfaction has been actively
studied since businesses shifted from the standard business model due to technological
disruption. Social networks and advanced technologies were successfully integrated into almost
every sphere of everyday life. This paper aims to understand the phenomenon of Instagram - one
of the most favoured social networks that had a great influence on the B2C business model in
Kazakhstan. Instagram was launched in 2010 and within only 10 years has fully changed lives.
Instagram uses data mining for two main purposes: (i) Instagram as a company: a social network
using big data to increase the customer (user) satisfaction rate and number of users and (ii)
Instagram as a platform: profiles (business and private) using the Instagram algorithms and big
data to interact with customers and increase the customer satisfaction rate.</p>
      <p>These two topics will be studied and researched based on the available information and
statistics and previous studies of the related issues using the case study methodology. Case study
is one of the most popular research methods in the business field. The case study allows
investigating the specific issues within the boundaries of the particular example of the individual,
situation or organization. In this research, the topic of data mining to increase customer
satisfaction will be studied based on the example of a huge social network - Instagram. An
indepth analysis of the Instagram phenomenon will be investigated in a real-life context.</p>
      <p>As it was mentioned above this social network uses big data to successfully fulfill two roles
as a Company itself and as a platform for other companies. During the past 10 years, technological
innovations stepped a lot forward. Also, Instagram consumers' (users) preferences shifted to the
inclination to see more personalized content. Instagram created a new business model making
companies shift from internet website platforms to social media profiles. If 5 years ago most of
the content in this social network had an entertaining context, now it serves as a business,
educational, and informational platform. Instagram uses two types of analysis: content-based and
structure-based. Content-based analysis is based on content posted by a particular person
(images, videos, text), and his/her responses to the content posted by other people (likes, reposts,
comments). The structure-based analysis is used to track geolocations, social links, and
communities based on users' internet behaviour without any obvious clicks.</p>
      <p>Instagram closely monitors its users' behaviours and reactions. It constantly updates an
application with new features based on the audience's interests and needs. During the pandemic
in Spring 2020 new feature of "Co-Watching" was added that allowed people to share posts over
video calls as many people stayed at home and needed to share their leisure time with their
families and friends.</p>
      <p>It is hard to imagine any company that does not have a business account on Instagram as it
provides a variety of useful tools and functions for the owners to get the full picture of the target
audience, their profile, preferences and behaviour. As described in the picture below there is a
constant flow of personal data, publicly available data and personalized offerings and ads. System
infrastructure represents several data servers, 3rd party applications, user profiles and
interconnected search engines.</p>
      <p>The community of Instagram users in Kazakhstan is one of the largest worldwide. There are a
lot of small and medium enterprises that fully base part of their business processes including
sales, marketing, customer relationship management, and inventory management on this social
platform. The application provides businesses with great opportunities to save money on data
gathering as it is already available with real-time statistics in the context of each post, profile and
product.</p>
      <p>Instagram stores big amounts of data and in this research, authors aimed to identify the effect
of social media data mining on customer satisfaction. To understand the issue, both qualitative
and quantitative research approaches are used for in-depth insights into the case related to the
research topic. The qualitative approach relies on data received from the primary sources. Data
for the research purposes will be gathered through first-hand observations and interviews with
individuals identified as the target audience interested in the case. Observations provide an
understanding of the situation and change in the circumstances and conditions and show the
direct relationship based on user experience. Interviews with representatives of different
audiences allow for receiving an opinion from various angles. The type of quantitative approach
to data gathering used is a questionnaire with closed questions where respondents can only
choose their answer from the suggested options. Considering the time limits, the questionnaire is
the most suitable way to reach a greater number of users in a shorter amount of time and at zero
cost. Results will be provided in a structured format and then will be transformed into statistical
and graphical data for visualization and representation of users' opinions.</p>
      <p>Research methodology is developed in such a way that the main idea described by the
Instagram case study supporting the fact that customer satisfaction is quite dependent on the
quality and amounts of data mined from social media is further investigated and studied by the
researchers in the context of Kazakhstan. Data collected from the primary sources will give
opportunities for further research, showing if the opinions are challenging or supporting the case
idea.</p>
    </sec>
    <sec id="sec-4">
      <title>4. Research design</title>
      <p>The main objective of the research is to find out if Instagram uses i) data mining to increase the
customer (user) satisfaction rate and number of users from the perspective of a social media
company and ii) the Instagram algorithms and big data on Instagram platform to interact with
customers and increase the customer satisfaction rate for business and private users.</p>
      <p>
        The chosen research method combines qualitative and quantitative approaches. The goal of a
mixed approach is to expand and strengthen the research's conclusion. The element of the
qualitative method is conducting the interviews, while the quantitative one is a questionnaire.
Also, the case study was applied [
        <xref ref-type="bibr" rid="ref17">17</xref>
        ].
      </p>
      <p>The quantitative method is usually applied for acquiring statistical data, numerical data from
specifically designed surveys, questionnaires and other structured research instruments. These
research instruments are accurately designed for the representatives of certain target
populations. The target audience is initially set by the research team. A target audience is a
defined group of people with common criteria, demographics, location or other factors. Generally,
representatives of the target audience are chosen randomly. Thus, the answers collected are
assumed not to be biased. Selecting the target audience is one of the key points for useful data for
the research, which is why this step requires an accurate assessment from the research team.
Before launching the designed questionnaire or survey for the target audience, it should be tested
by the research team, to be sure that the target audience does not face any challenges related to
understanding and completing the questionnaire. Using the questionnaire tool is helpful for its
rapidity, meaning that the results are obtained very quickly in a short period, without any tough
efforts for both the research team and the target audience. Moreover, there is an opportunity to
cover a wide range of representatives from the target audience without any significant costs.</p>
      <p>In this research, the research team chose an online questionnaire as a structural research
instrument. Questionnaires were designed to understand if the Instagram algorithm adapts to
customer needs and values and raises customer satisfaction and interaction. The questionnaire
was designed via Google Forms. During the process of designing the survey instrument for the
research, the main goal was to keep it simple, not to use technical terminology, not to add too
many questions, but to collect all the necessary information required. the team managed to
expand the number of representatives from the target audience. The target audience is users of
Instagram social media who are age 16+, and living in Kazakhstan. The research team designed
the questionnaire to be anonymous. The reason for anonymous questionnaires is to inspire
responders to disclose sensitive information. The questionnaire was published on the Instagram
and LinkedIn profiles of every member of the research team, to make it publicly available. The
total number of respondents is 106. Results are described in the Research design part, and the
questions designed for the questionnaire are attached in the appendix of this research paper.</p>
      <p>A qualitative method is an approach that is aimed at acquiring results via open-ended
questions and conversational communication. Qualitative methods assume using in-depth
interviews, case studies, content analysis, focus groups, ethnographic research, and observation.
The qualitative method allows the research team to acquire a non-numeric and more detailed
view of the target audience. To ensure that qualitative data acquired makes a valuable input for
future findings, the data sources should vary.</p>
      <p>This research team chose in-depth interviews and case studies as methods of qualitative
approach. In-depth interview methods give more detailed findings in selected research topics.</p>
      <p>Next, Instagram as a social media company was selected as a case study approach. A case study
is one of the methods of the qualitative approach. The success of Instagram as a social media
company from the perspective of customizing and increasing customer satisfaction is a major
research goal. Since the research team selected two data sources for a qualitative approach, it is
assumed the results to be accurate, precise and reliable.</p>
    </sec>
    <sec id="sec-5">
      <title>5. Research results and analysis</title>
      <p>The results of quantitative research based on collected primary data from questionnaires are
described below.</p>
      <p>Figure 1 shows the results of the percentage of different types of accounts represented in the
population sample of the research. The proportions do not describe the real picture as the sample
size was collected from the followers of the research team.</p>
      <p>The information in Figure 2 shows the frequency of purchases made by respondents on
Instagram. 71% of users from the sample use Instagram to buy something. 65% of respondents
stated that Instagram advertisement (including both target and influencer ads) affects their
decision to buy the product. 52% of individuals purchased an ad of the product demonstrated on
the Influencer's account. 95% noticed that Instagram offers ads for a particular product after they
discussed it or one of its complementary goods loudly during real-life conversations.</p>
      <p>Respondents were asked about the aspects of Instagram they like and dislike the most. The
top 3 features people enjoy about Instagram are the content itself (including stories, posts, and
recommendations), the opportunity for socializing (for example, chatting with friends and
sharing information with them) and the convenience of the interface (simplicity of use and
functions). 3 main drawbacks of using this social network according to the respondents' opinion
are advertisements (too high frequency, sometimes the ad is useless for the particular person in
a given period), worthless content (not unique, or content without an interesting topic or with
low quality) and social issue (mainly including the problem of insincerity). Data protection issue
concerns the users of Instagram. 59% of respondents answered that they think their personal and
behavioural data is in danger because Instagram collects it to use for further algorithms.</p>
      <p>The sample was asked to rate their satisfaction level with the services provided by Instagram.
Figure 4 visualizes the results, as it is shown no one indicated his satisfaction level as low.</p>
      <p>Figure 4 shows the degree to which Instagram users believe they are satisfied with the social
network.
online for their offline hobbies or free time, 10% visit accounts for education purposes, while only
7,5% consider Instagram only for entertainment.</p>
      <p>98% out of 106 respondents will remain Instagram users in the future and they have no intent
to stop using it or shift to the other network.</p>
      <p>The main outcomes from interviews are demonstrated below, they show the results of the
qualitative research.</p>
      <p>Instagram Influencer</p>
      <p>According to Influencers' opinion, Instagram is a great platform for self-promotion. They state
that for so many years of being an Instagram bloggers, they had collected the audience which is
inspired by their Instagram web pages. In turn, Instagram bloggers are happy to share with
people their daily lives, products and services that satisfy their expectations and values. Also, they
get a stable income from being Instagram bloggers. With the appearance of Instagram, a new term
"Influencer" emerged and it allowed them to start to earn more money. Being an Influencer is a
real job, it takes time, resources and energy. However, the further the more difficult it is for people
to become an influencer, because of high competitiveness. Simple lifestyle content is not enough
to become a successful blogger - consumers want something new, creative and interesting.</p>
      <p>Instagram business account owner</p>
      <p>Given the fact that Instagram has become one of the most popular e-commerce platforms,
more and more businesses enter this marketplace. Instagram is affordable for selling and
promoting both goods and services for various companies. There is a possibility to create an
Instagram account for the small firm and sell effectively on this platform without a physical store
location and with a limited budget. Instagram allows information about followers and gives the
full statistics. Using this data businesses can better understand their customers and with the use
of appropriate marketing tools affect their satisfaction level. What is more important is that with
the help of Instagram, it is possible to launch target advertisements which are cheaper than, for
example, TV ads. Target ads make it possible to spend money to attract a particular group of
buyers rather than waste the budget and show an ad to a huge uninterested pool of people. First
of all, business page owners consider Instagram as the most effective platform for target reach
and data analysis.</p>
      <p>End-User</p>
      <p>Instagram 10 years ago and Instagram today are two different platforms - everything has
changed starting from the design to the overall concept. It seems that now this social network is
more about commerce. It is easy to search for needed products on Instagram by hashtags, geotags,
and company account names. All the information is there including photos, videos, and feedback
from other clients - this is convenient. Search engines, such as Google are used less for shopping
purposes because Instagram is more fun, it integrates shopping and entertainment. Instagram
ads are another interesting tool as they show exactly what was talked about or written about in
a message. It makes life easier and faster by offering the goods that are needed promptly.
However, when the advertisement appears too frequently it becomes annoying. Especially when
ads were just for introduction. It is also a pleasure to observe incremental changes and
improvements in interface and functionality from time to time.</p>
      <p>Marketing Specialist of corporation</p>
      <p>Huge corporations with experienced sales and marketing teams, extended budgets and aligned
strategies use Instagram accounts not just to sell their product but also to build effective
communication with consumers. Instagram is used mostly as a data collection tool, rather than a
marketplace. Geospatial analytical data is widely used by retail companies to organize customer
journeys online and offline. Information about clicks, reactions, and comments is used not only to
promote products on Instagram but also to design new products or make changes to existing
ones. Large businesses, like retail use various types of advertisement including target, opinion
leaders, sponsorship and collaboration to increase the visibility of products. More and more firms
understand that people spend all their lives on social networks and are trying to increase their
social media presence. Instagram encourages companies to enter it by creating user-friendly
accounts and functionality. With the help of Instagram statistics sales are increasing because
companies can understand their customers directly.</p>
    </sec>
    <sec id="sec-6">
      <title>6. Research discussions and analysis</title>
      <p>“Instagram isn't necessarily a photo company or a communications company as I like to say; we're
also going to be a big data company" - As quoted by Kevin Systrom, CEO of Instagram. With
Instagram's exponentially growing popularity and ever-growing number of users, it's only fair to
ask what is their secret. We say most of it comes from its strategic use of big data.</p>
      <p>By analyzing both qualitative and quantitative methods' results and the reviewed literature,
the research team got a clear finding regarding two main research goals: i) implementing data
mining to increase user satisfaction and the number of users from the perspective of a social
media company and ii) the Instagram algorithms and big data usage and processing on the
Instagram platform to interact with customers and increase the customer satisfaction rate for
business and private users.</p>
      <p>The research team made the following assumption for further observation: Does Instagram
express itself as a big data company and how they use data that millions of its users around the
world generate daily to create personalized feeds and explore pages that users will most likely
"like", target ads that users will most likely click as well as bots and systems that clear spams and
block sensitive contents or comments further increasing customer satisfaction among users. Also,
the research questions covered the effectiveness of Instagram features for reusing its vast data
and finding a way to be beneficial to both businesses and end customers.</p>
      <p>Finalizing the results of quantitative research, we can see Instagram does affect buying
patterns due to its personalized advertisements and has very high user satisfaction level. The
research team found that 65% of respondents stated that Instagram advertisements (including
both target and influencer ads) affect their decision to buy the product. 52% of individuals
purchased an ad of the product demonstrated on the Influencers’ accounts page. The most
interesting part is that almost all the respondents said that the Ads were not some random
products but a specific thing that the user previously discussed or looked on the internet to find
or somehow was interested in. As a result of implementing this algorithm, Instagram gives each
user a very personal and specific experience. Using data mining tools Instagram identifies each
user and creates a specific portfolio for each one of their users. By analyzing liked posts, visited
pages, searched things, visited places, used hashtags, followers, and customers' demographic
factors, Instagram offers very customized content. Personalized ads and customized
recommendations make each customer's Instagram usage experience unique. As a result,
customer satisfaction rises and makes them want to come back and get more offers. One of the
things mentioned by respondents is that they don't like the fact that usage of Instagram leads to
addiction. Once a user wants to check his page, he stays online for hours scrolling the feeds and
suggested publications. The management of individuals gets worse, but on the other hand,
business accounts make more money from increased visitor numbers. The more time people
spend in this social network, the more engaged they are. Consequently, high engagement is equal
to a high satisfaction level. Users that consider Instagram as an interesting platform and have a
positive perception of it, tend to be easily influenced and attracted by promotions, ads and other
offerings.</p>
      <p>To sum up the research results, the first major fact claimed is that Instagram as a company
does collect personal data to offer customers very personalized content. As it turned up from the
questionnaire results, this point is the main Instagram advantage to maintain and enhance
customer satisfaction. This is applicable not only for entertaining content and chatting but also
for learning different courses, news reading and other educational purposes. The majority of end
users confess that they like recommendations based on data acquired from them, such as likes,
followers, clicks etc. On the other hand, people are concerned about the confidentiality issue.
Hopefully, Instagram will also follow the ethical standard of using personal data and pay close
attention to the degree of personal information they acquire and use it for their benefit. An
example of an ethical issue raised by their related party Facebook, which faced legal claims
regarding sharing people's data with third-party companies in 2019.</p>
      <p>The second major finding is that the Instagram platform is a helpful and impactful tool for
business accounts and marketing specialists from the industries. According to the result of the
indepth interview, there is a great opportunity for small business accounts to promote the business
online without significant costs, for example, rent expenses. The major finding from both
qualitative and quantitative methods is that Instagram collects data for further marketing
assumptions. There is an opportunity for big corporations to acquire data, set correct target
audiences and promote their services. Moreover, it is proven by questionnaire results that
majorly Instagram algorithm recommends promoted services and products to the right audience.
This effect is achieved mostly by the "hearing" function of the algorithm when products appear in
ads after discussing them in real life. On the other side, the research team found out that
sometimes ads do not appear for the right audience, and it annoys end users. Thus, Instagram
should update the algorithm to a more sensitive one to increase and maintain customer
satisfaction.</p>
      <p>Imagine that your company has great expertise in big data management that, in a couple of
years, turns into something more - Customer Value Management (CVM) and truly big data. This
is exactly what happened with Instagram: they learned how to manage churn, effectively predict
the sales of products and conduct campaigns to promote and offer personalized offers to each
client.</p>
      <p>This expertise and data analytics consider hundreds of parameters related to subscribers:
their socio-demographic profiles from gender and age to region of residence, number of family
members and personal interests. The same expertise requires maximum personalization of any
message, contact with a subscriber, and an exclusive offer here and now. And Instagram is one of
the best examples of usage of such analytics. The research team believes that if Instagram follows
its data-driven strategy, it will enhance the company's leading position.</p>
    </sec>
    <sec id="sec-7">
      <title>7. Research conclusions</title>
      <p>Today's customers want companies to treat them as unique individuals, anticipate their needs
and wants, remember their peculiarities and preferences and know their purchase history. All
this has a positive effect on user and customer experience and as a result, leads to increased
customer satisfaction. The era of consumer society ended several years ago and brand
significance level declined with time. Companies are now focused on a personalized approach to
each client, tracking the history of purchases and product recommendations based on experience.
According to the customers' opinion, they are ready to provide their historical and geospatial data
if it will increase their overall experience with the company and create more relevant content.</p>
      <p>Instagram is one of the most successful companies actively using big data to increase the
satisfaction of its users (individuals and businesses) as well as providing other companies with
an opportunity to use customer data for business development purposes. This is the main idea of
the research case study which is supported by the results received from surveys and interviews.
Time spent on Instagram is increasing due to content recommended based on personal interests.
This means that this platform is satisfying the needs of its users as an entertainment platform.
However, at the same time, this social network managed to realize itself as an e-commerce
platform. Results show that 8 out of 10 people use Instagram to buy something at least once in a
couple of months. In Kazakhstan, Instagram became an effective platform for paid advertisement
as well as organic content. This network helps foster intercommunication and interactions with
brands and companies due to its well-designed functionality. 98% of users tend to continue using
this social platform which is an extremely good indicator of user loyalty.</p>
      <p>Research results could provide deeper insights if the number of respondents is increased up
to 1000 with the diverse sample including a greater number of business accounts and bloggers.
Also, it might be interesting to see the sphere of each business profile. Spending more time
researching the topic would allow going through each Instagram metric and using an
experimental approach (i.e., checking the quality of content offered in the result of audio mining).</p>
      <p>Another point of the research that might need more investigation of the issue is data privacy.
Almost half of the users are concerned about the collection and usage of their data and the fact
that global networks store data about all movements and interactions bothers many people.</p>
      <p>Research provides valuable data on how different types of Instagram users perceive data
mining and how it influences their interest in the businesses and the social network itself.</p>
    </sec>
    <sec id="sec-8">
      <title>8. References</title>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <surname>Enberg</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          (
          <year>2021</year>
          ).
          <source>Social Media Update. Insider Intelligence</source>
          ,
          <volume>1</volume>
          (
          <issue>1</issue>
          ), 26. https://www.businessinsider.com
          <article-title>/global-social-network-users-report.</article-title>
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <surname>Pushpam</surname>
            ,
            <given-names>C.A.</given-names>
          </string-name>
          , &amp;
          <string-name>
            <surname>Jayanthi</surname>
            ,
            <given-names>J.G.</given-names>
          </string-name>
          (
          <year>2017</year>
          ).
          <article-title>Overview of Data Mining in Social Media</article-title>
          .
          <source>International Journal of Computer Sciences and Engineering</source>
          ,
          <volume>5</volume>
          (
          <issue>11</issue>
          ),
          <fpage>147</fpage>
          -
          <lpage>157</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <surname>Handley</surname>
            ,
            <given-names>L.</given-names>
          </string-name>
          (
          <year>2019</year>
          , January 24).
          <article-title>three-quarters of the world will use just their smartphones to access the internet by 2025</article-title>
          . CNBC. https://www.cnbc.com/
          <year>2019</year>
          /01/24/smartphones72percent-of
          <article-title>-people-will-use-only-mobile-for-internet-by-2025.html</article-title>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          [4]
          <string-name>
            <surname>Sudarshan</surname>
            ,
            <given-names>V. R.</given-names>
          </string-name>
          (
          <year>2017</year>
          ).
          <article-title>A Data Mining Approach to Modeling Customer Preference: A Case Study of Intel Corporation</article-title>
          . ARIZONA STATE UNIVERSITY,
          <volume>1</volume>
          (
          <issue>1</issue>
          ), 1, https://repository.asu.edu/attachments/194145/content/Ram_asu_0010N_
          <fpage>17584</fpage>
          .pdf.
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <surname>Shmueli</surname>
            ,
            <given-names>G.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Bruce</surname>
            ,
            <given-names>P. C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Yahav</surname>
            ,
            <given-names>I.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Nitin</surname>
            ,
            <given-names>P. R.</given-names>
          </string-name>
          , &amp;
          <string-name>
            <surname>Jr. Lichtendahl</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          (
          <year>2018</year>
          ). Data.
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <surname>TAMA</surname>
            ,
            <given-names>B. A.</given-names>
          </string-name>
          (
          <year>2015</year>
          , May 10).
          <article-title>DATA MINING FOR PREDICTING CUSTOMER SATISFACTION IN FAST-FOOD RESTAURANTS</article-title>
          .
          <source>Journal of Theoretical and Applied Information Technology</source>
          ,
          <volume>75</volume>
          (
          <issue>1</issue>
          ), 19. https://www.jatit.org/volumes/Vol75No1/3Vol75No1.pdf.
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <surname>Hassani</surname>
            ,
            <given-names>H.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Huang</surname>
            ,
            <given-names>X.</given-names>
          </string-name>
          , &amp;
          <string-name>
            <surname>Silva</surname>
            ,
            <given-names>E.</given-names>
          </string-name>
          (
          <year>2018</year>
          , June 27).
          <article-title>Digitalisation and Big Data Mining in Banking</article-title>
          .
          <source>Big Data Cognitive Computing</source>
          ,
          <volume>2</volume>
          (
          <issue>3</issue>
          ), 20. https://www.mdpi.com/2504- 2289/2/3/18/htm.
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <surname>Chen</surname>
            ,
            <given-names>D.</given-names>
          </string-name>
          ,
          <string-name>
            <given-names>Laing</given-names>
            <surname>Sain</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            , &amp;
            <surname>Guo</surname>
          </string-name>
          ,
          <string-name>
            <surname>K.</surname>
          </string-name>
          (
          <year>2012</year>
          ,
          <article-title>August 27)</article-title>
          .
          <article-title>Data mining for the online retail industry: A case study of RFM model-based customer segmentation using data mining</article-title>
          .
          <source>Journal of Database Marketing &amp; Customer Strategy</source>
          Management volume,
          <volume>19</volume>
          (
          <issue>1</issue>
          ),
          <fpage>197</fpage>
          -
          <lpage>208</lpage>
          . https://link.springer.com/article/10.1057/dbm.
          <year>2012</year>
          .
          <volume>17</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <surname>He</surname>
            ,
            <given-names>W.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Tian</surname>
            ,
            <given-names>X.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Tao</surname>
            ,
            <given-names>R.</given-names>
          </string-name>
          , Zhang,
          <string-name>
            <given-names>W.</given-names>
            ,
            <surname>Yan</surname>
          </string-name>
          ,
          <string-name>
            <given-names>G.</given-names>
            , &amp;
            <surname>Akula</surname>
          </string-name>
          ,
          <string-name>
            <surname>V.</surname>
          </string-name>
          (
          <year>2017</year>
          , November 13).
          <source>Online Information Review. Emerald Insight</source>
          ,
          <volume>41</volume>
          (
          <issue>7</issue>
          ),
          <fpage>1468</fpage>
          -
          <lpage>4527</lpage>
          . https://www.emerald.com/insight/content/doi/10.1108/OIR-07-2016-0201/full/html.
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <surname>Kushnazarov</surname>
            ,
            <given-names>F.</given-names>
          </string-name>
          (
          <year>2019</year>
          ).
          <article-title>Consumer Life Cycle and Profiling: A Data Mining Perspective</article-title>
          . IntechOpen. DOI: http://dx.doi.org/10.5772/intechopen.85407.
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [11]
          <string-name>
            <surname>Hassan</surname>
            ,
            <given-names>M. M.</given-names>
          </string-name>
          , &amp;
          <string-name>
            <surname>Tabasum</surname>
            <given-names>M.</given-names>
          </string-name>
          (
          <year>2018</year>
          ).
          <article-title>CUSTOMER PROFILING AND SEGMENTATION IN RETAIL BANKS USING DATA MINING TECHNIQUES</article-title>
          .
          <source>International Journal of Advanced Research in Computer Science</source>
          ,
          <volume>9</volume>
          (
          <issue>4</issue>
          ),
          <fpage>24</fpage>
          -
          <lpage>29</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [12]
          <string-name>
            <given-names>A.</given-names>
            <surname>Salloum</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            ,
            <surname>Al-Emran</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            ,
            <surname>Monem</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A. A.</given-names>
            , &amp;
            <surname>Shaalan</surname>
          </string-name>
          ,
          <string-name>
            <surname>K.</surname>
          </string-name>
          (
          <year>2017</year>
          ).
          <article-title>A Survey of Text Mining in Social Media: Facebook and Twitter Perspectives</article-title>
          .
          <source>Advances in Science, Technology and Engineering Systems Journal</source>
          ,
          <volume>2</volume>
          (
          <issue>1</issue>
          ),
          <fpage>127</fpage>
          -
          <lpage>133</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          [13]
          <string-name>
            <surname>Boonjing</surname>
            ,
            <given-names>V.</given-names>
          </string-name>
          (
          <year>2017</year>
          ).
          <article-title>Data Mining for Positive Customer Reaction to Advertising in Social Media</article-title>
          .
          <source>Information Technology for Management. Ongoing Research and Development</source>
          ,
          <volume>83</volume>
          -
          <fpage>95</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          [14]
          <string-name>
            <surname>Wang</surname>
            ,
            <given-names>Y.</given-names>
          </string-name>
          (
          <year>2020</year>
          ).
          <article-title>Unpacking the impact of social media analytics on customer satisfaction: do external stakeholder characteristics matter</article-title>
          ?
          <source>International Journal of Operations &amp; Production Management</source>
          ,
          <volume>40</volume>
          (
          <issue>5</issue>
          ),
          <fpage>647</fpage>
          -
          <lpage>669</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          [15]
          <string-name>
            <surname>McCourt</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          (
          <year>2018</year>
          ).
          <article-title>Social Media Mining: The Effects of Big Data In the Age of Social Media. Yale Law School or the Media Freedom and Information Access Clinic</article-title>
          .
        </mixed-citation>
      </ref>
      <ref id="ref16">
        <mixed-citation>
          [16]
          <string-name>
            <surname>Mohamed</surname>
            <given-names>A.</given-names>
          </string-name>
          <string-name>
            <surname>Hamada</surname>
          </string-name>
          , Adejor E.
          <string-name>
            <surname>Abiche</surname>
          </string-name>
          (
          <year>2022</year>
          ), “
          <article-title>Adopting Data Mining and Social Media Analytics to Achieve Customer Satisfaction”</article-title>
          ,
          <source>7th International Conference on Digital Technologies in Education, Science and Industry (DTESI</source>
          <year>2022</year>
          ), Vol.
          <volume>3382</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref17">
        <mixed-citation>
          [17]
          <string-name>
            <surname>Brannen</surname>
            <given-names>J</given-names>
          </string-name>
          , editor. “
          <article-title>Mixing methods: Qualitative and quantitative research”</article-title>
          .
          <source>Routledge; 2017 Jul</source>
          <volume>12</volume>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>