You are on page 1of 4

Cloud and Big Data Assignment – 2

Literature Review
Group 14
Roll Number Name PRN

43108 Adarsh Vaish 20020141008

43208 Ashutosh Kharsan 20020141108

43226 Nandan Chouda 20020141126

43246 Pranali Ingle 20020141146

43261 Rajvi Agrawal 20020141161

43288 Shivank Sharma 20020141188

43297 Shubham Khaitan 20020141197

43298 Shubham Nagpal 20020141198

43302 Siddhant Agrawal 20020141202

43306 Sneha Agrawal 20020141206


Data Mining:
Data mining is an artificial intelligence branch that focuses on extracting information from
real-world data sets. It's an interesting interdisciplinary field that includes, among other
things, machine learning, statistics, pattern recognition, databases, and information
retrieval.

Data Mining in Retail:


Companies in the retail sector are constantly aware of the need to gather data and turn it
into actionable insights into innovative processes, products, and services. To close the
horizontal and vertical gaps between supply chain stakeholders, it's critical to understand
data from various sources. It's also crucial to know how consumers feel when they pass
across the supply chain. CRM (Customer Relationship Management) systems are now widely
used to manage, sustain, and improve customer relationships. There is a slew of other
analytics tools geared toward attracting and holding customers. Customers who are
churners, or customers who have a high risk of terminating their partnership with the
company, must be detected in order for these resources to be useful. CRM systems will
segment consumers based on their loyalty to the business in this way, helping them to make
better decisions. The paper uses a descriptive analysis accompanied by a predictive analysis
to establish a step-by-step guide for predicting turnover in a retail organization. The first
step in the approach outlined in the paper is to describe churn in the sense of a grocery
store. The paper's second section collects actual data from retail businesses. The paper
examines the factors that impact churn using descriptive analysis. In this way, a model will
learn about a customer's consumption habits. The third and final component uses CRISP-DM
methods to build a predictive model. CRM applications are often used to produce a churn
predictive model that is reliable.

Data mining techniques for product and price optimization, such as predictive systems and
recommendation engines, will become more popular in the future, allowing for fully
automated interactions with customers in stores. Interactivity and adaptability will be
emphasized, with online learning playing a key role. Another pattern is the management of
the customer life cycle and the maximization of the customer life time value. To a large
degree, this is technically possible today; all data for the customer life cycle is theoretically
present somewhere in the company; however, they are seldom used in this way because
they are not fully integrated into the IT-process.
The process of dividing a market into distinct segments based on certain characteristics is
referred to as "segmentation." Data mining may be used to group or cluster customers
based on their operation. This information can be used to build clusters of similar
customers, retain good customers, and classify likely responders for targeted marketing.
Many businesses have already begun to use data mining technology to gain a strategic
advantage over their competitors. In the manufacturing industry, data mining allows for a
better understanding of procedures, as well as the detection of device failures and
preventive maintenance. As a result, data mining assisted in improving efficiency and
lowering costs associated with damages or lack of prediction.
Data Mining in Healthcare:
In the healthcare sector, there are patients and their conditions, health insurance
companies, medication and treatment schedules for different illnesses, emergency
departments, and so on. The amount of data available is immense. To discover secret
healthcare facts and trends, a variety of data mining techniques can be used.
There are still areas that are attempting to explore data mining and put it to good use, one
of which is the healthcare and medicine industry There is a large amount of data stored in
real-world databases, and it is constantly increasing, creating both an incentive and a need
for semi-automatic methods to find out what is known in those databases. If the results are
successful, they can be used to enhance an organization's decision-making process. For
example, patient data may reveal previously unknown information about which patients are
more likely to develop a particular disease. This awareness could help potential patients get
a better diagnosis and treatment. In this paper, the Arusha area is investigated in order to
incorporate data mining as a substitute. Data mining technology should be implemented
and used in the Arusha area, according to the authors, because it will boost the healthcare
industry and residents' well-being. The incorporation of data and text mining, using digital
diagnostic images that can be carried into healthcare data mining applications, is proposed
in the paper. The terms data mining and information exploration are not synonymous.
The author acknowledges the opposition to data mining among doctors and the medical
community, and lists a number of factors that can benefit the industry.
Data mining can be used to assess the effectiveness of a strategy. For example, data mining
may help find the right treatment for a patient by examining their treatment history,
symptoms of various diseases, and comparing causes. A group of patients with the same
illness or disease who have already been treated with different drug regimens can be
compared to determine the most cost-effective yet best treatment. It would also allow
physicians to compare their practise practises to those of others, as well as peer-reviewed
industry guidelines. In this way, data mining would allow for efficient standardisation.
Data overload in old databases makes it difficult to analyse and research data effectively;
however, incorporating data mining will assist doctors in practising evidence-based data
medicine and preventing hospital errors. Data mining can assist in monitoring the spread of
diseases as well as deaths caused by different diseases, which can be useful in developing
policies for people. Tracking a patient's progress as well as his or her family's medical
background may aid in the treatment or prevention of major diseases. Data mining can also
help avoid pandemics by monitoring and diagnosing patients in a non-evasive manner.
It can be used to enhance high-risk disease identification, establish successful therapies, and
monitor chronic disease states. Furthermore, data mining methods may be used to reduce
the number of hospital admissions and claims. Various medical centres are using data
mining strategies to shorten hospital stays, minimise health risks, optimise medical
practises, improve patient quality, and provide information to physicians. These will improve
healthcare quality while remaining cost-effective in the long run.
Medical data fraud and misuse can be expensive, and they're also related to other issues,
such as personal information. Data mining methods may be used to find these. Data mining
techniques, for example, may be used to detect odd or unusual patterns of claims from labs,
hospitals, physicians, or any other source. Inappropriate drugs, referrals, and medical claims
from insurance providers can all be found using data mining software.
References:
https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.302.943&rep=rep1&type=pdf

https://www.researchgate.net/publication/22592742

https://run.unl.pt/handle/10362/112895

http://www.ijceronline.com/papers/Vol3_issue8/part%202/I0382073077.pdf
http://mines.humanoriented.com/classes/2010/fall/csci568/papers/Data_Mining_Health.pdf

https://www.researchgate.net/publication/321747011_Applying_Data_Mining_Techniques_i
n_Healthcare
https://www.researchgate.net/publication/322754945_DATA_MINING_IN_HEALTHCARE

You might also like