Problems and Solutions of Data Analytics in Business

Liwei Liu

School of Engineering and Applied Science, Gonzaga University, Spokane, U.S.A.

Keywords: Data Quality, Data Breaches, Business Data Analysis, Data Encryption, Interdepartmental Collaboration.

Abstract: Nowadays, business cannot thrive without data, and naturally, data analytics has become an indispensable

tool in business. Data analysis can not only intuitively provide decision-making support, but also help

enterprises to understand the market and customers more deeply and provide enterprises with more accurate

and efficient programs and development direction. But while data analytics in business provides results for

companies, there are also not a few problems. This study analyzes the problems of data analytics in business

by analyzing them and proposing countermeasures. This paper examines four problems of data analysis in

enterprises: data quality problem, data leakage problem, inaccurate data analysis problem, and data

application problem. In response to these problems, this paper establishes a data processing mechanism,

encrypts data, upgrades the tools and related talents for data analysis, and promotes good communication

among departments, so as to ensure the stability of data quality, prevent the leakage of information, improve

the accuracy rate, and enable better application in reality.

1 INTRODUCTION

In today's competitive and ever-changing business

environment, data analytics has become a critical

success factor for organizations. With the

advancement of digital transformation, companies

can use data analytics to get the results they want to

make strategic decisions, optimize operational

processes, and enhance customer experience,

ultimately determining the direction of the company.

Whether it's market trend forecasting, customer

information analysis, or supply chain optimization,

data analytics is driving companies to achieve more

efficient and competitive business models (Wang,

2024). These phenomena show that data analytics has

never been more important in business.

With the rapid advancement of data analytics,

organizations have undergone a dramatic

transformation in their ability to collect, store, process

and apply data. For starters, the rise of big data

technologies has enabled organizations to handle

unprecedented scales of data and extract valuable

information from it. For example, Amazon has used

its powerful data analytics to boost sales through

personalized recommendation systems (See Figure 1)

(Huang, Jiang, Wu, & Wang, 2020).

https://orcid.org/0009-0007-5621-2765

Second, the application of machine learning and

deep learning has enabled companies to make more

accurate predictions and decisions. These

technologies can automate the processing of large

amounts of complex data and identify potential

patterns and trends, thus helping enterprises gain an

advantage in the fierce market competition. For

example, an innovative decision support model

“AlphaVision” designed for stock price prediction by

seamlessly integrating real-time news updates and

Return on Investment (ROI) values, utilizing various

machine learning and deep learning approaches

(Divyashree et al., 2024).

However, while data analytics technologies

present tremendous opportunities, they are also

accompanied by a new set of challenges and

problems. In the face of these rapidly changing

technological and market environments, in-depth

analysis and optimization of data analytics

technology has become inevitable. Enterprises must

not only rely on existing technical means, but also

need to comprehensively improve the efficiency and

accuracy of data analysis from multiple dimensions,

such as data quality, data security, analytical tools and

talents, as well as cross-departmental cooperation.

Only in this way can enterprises maintain a leading

position in the future market competition and

Liu, L.

Problems and Solutions of Data Analytics in Business.

DOI: 10.5220/0013207700004568

In Proceedings of the 1st International Conference on E-commerce and Artiﬁcial Intelligence (ECAI 2024), pages 123-127

ISBN: 978-989-758-726-9

123

Figure 1: Example of Amazon.

continue to realize business growth. Therefore, this

paper will deeply explore the key issues of data

analytics in business applications and put forward

corresponding countermeasures, with a view to

providing valuable reference and guidance for

enterprises in the data-driven era.

2 PROBLEMS

2.1 Data Quality

In this era of information explosion, data and

information from a variety of sources, data types, so

when this paper is collecting data there will be

incomplete data, data duplication, and even incorrect

data. Handle and clean the data is one of the perennial

challenges in data analytics (Chu, Ilyas, Krishnan, &

Wang, 2016). Therefore, the quality of data is a key

determinant of the quality of decisions made

(Choughri, et al., 2018). When the data is duplicated,

it will cause great errors in data analysis. In life it is

also very easy to have data entry errors that lead to

incorrect data. When this paper unknowingly enters

incorrect information, it can make it difficult to find

effective ways to refine and modify the model. In

daily life, data entry errors are also a major cause of

incorrect data. For example, users may double-enter

information, or machines may make mistakes during

testing and recognition, all of which can make the

quality of the data much worse. And when this paper

unknowingly enters this incorrect information into

the system, this paper may be stuck in a predicament

where it is difficult to figure out effective ways to

refine and modify the model. These are all factors that

contribute to data quality problems, when the results

of the analysis may be biased and lead to poor

decision-making.

2.2 Data Leakage

As the amount of data increases, the issue of data

privacy and security becomes more and more

important. All that is needed for data analytics is a lot

of data, and the more data is piled up then the greater

the risk of leakage. Chen et al. (2017) discussed

relationships revealed in this study. “As the volume

of data is growing exponentially and data breaches

are happening more frequently than ever before,

detecting and preventing data loss has become one of

the most pressing security concerns for enterprises.”

. So, the more organizations use and store data, the

greater the risk of data leakage and unauthorized

access. Cyber attackers will invade the enterprise

network through malware, phishing attacks, DDoS

attacks and other means to steal sensitive data. As

technology advances, attackers' tactics become more

sophisticated, and traditional firewalls and anti-virus

software can no longer fully defend against these

advanced threats. In addition, corporate insiders may

also cause data leakage through intentional or

unintentional behavior. Insiders often have higher

privileges and direct access to data, so when they are

the source of a breach, the damage is often more

severe. A data breach can then lead to data analysis

results that may become inaccurate for company

decisions or even affect the advancement of data

analysis. Not only can this lead to financial losses, but

it can also have a serious impact on a company's

reputation.

2.3 Inaccurate Data Analysis

The business market is evolving rapidly, so the cost

of trial and error for companies is remarkably high.

So, the accuracy of data analysis is crucial for

companies. Due to the differences in the selection of

ECAI 2024 - International Conference on E-commerce and Artiﬁcial Intelligence

124

data analysis tools and the differences in the

professionals, so the problem of inaccuracy may

occur in the market and in the decision making of

companies using data analysis. When data analysts

are not technically literate enough, it may be difficult

to extract valid values from the data or perform more

complex and large data analysis. When the tools do

not match the actual needs, there is a high probability

of overfitting and inefficient analysis effects. The

impact of these data analysis inaccuracies on the

company and the market is huge. Once the error is

large it may lead to poor decision making and wrong

direction of development of the company, resulting in

irreversible consequences.

2.4 Application of Data Analytics

Even if the results and accuracy of the data analysis

are good, it is useless when the results of the analysis

are not well applied to the market and the company's

decisions and actions. If companies do not make good

use of data processing and utilization in the market,

then they will be eliminated by their competitors (Hu,

2018). Just like our human body, when our brain

receives the information and completes the

instructions, if our nerve center is not able to transfer

the instructions accurately and quickly to all the

organs of our body, then our body won't be able to

function properly. In reality, when a data analytics

scientist analyzes market and company data and gets

a result, and the business department is unable to

understand the result and distribute it to the executive

departments, then the result is not well applied in

decision making and action.

3 SOLUTIONS

3.1 Reasonable Processing of Data

For the problem of data quality, what this paper has

to do is to improve the quality of data, that is, to make

reasonable processing of data. This paper can carry

out data cleansing to clean up the missing values and

duplicate values in the data set. In the enterprise there

should be data analysis scientists according to the

different requirements of each project, to establish a

set of appropriate data cleaning mechanism, to help in

data analysis in the data analysis to improve and

optimize the data analysis, regular review of the data,

fill in the missing values, remove duplicate data, so as

to improve the quality of the data. For example, for

missing data, data preprocessing and missing value

verification are performed to ensure data integrity.

For duplicate data, data cleansing techniques are used

to remove duplicate data and ensure that each piece

of data appears only once in the data set. For incorrect

data, the end can be achieved by multiple data entry

and comparing the correctness. Moreover, the

company should also develop a unified standard and

process for data processing, so that whenever a new

data set is added to the analysis, it will be

standardized in this process, maintaining the integrity

and consistency of the data, so that the data can be

more accurately applied to data processing.

3.2 Encrypting the Data

To solve the data leakage problem, enterprises should

strengthen both external protection and internal

prevention. First of all, strict access policies and

management rights should be formulated so that only

core personnel or authorized personnel can access.

When accessing and processing important data, you

can let multiple top-ranking data analytics scientists

have some of the permissions, and only when the

majority of scientists are authorized can they legally

access it. In this way, even if there is a problem with

an insider, it will not affect the entire data leakage.

Second, the data can be strictly protected using

encryption so that even if the company's data is

hacked, it cannot be accessed illegally. This is

equivalent to provide a double insurance for the

protection of data, for example, this paper can use

irreversible encryption algorithm for data processing,

which can effectively prevent reverse decryption. Just

like the new encryption method of data shown in the

study by Eltayieb et al., (2024) the encryption method

of CLPRE-CRF is proposed to prevent data leakage

and enhance data security. Then this paper also needs

to strengthen the risk assessment of data reading to

prevent violent decryption to obtain data. Finally, an

inspection team can be established, specifically

responsible for data protection, to conduct regular

data security checks, identify and repair potential

security gaps, to ensure that data security measures

are always effective. Once a data leakage problem

occurs, it will be immediately investigated and solved

to ensure that the data leakage problem is solved at

the first time.

3.3 Elevation Tools, Talents and Models

To improve the accuracy of data analysis, it is

necessary to improve both tools and talents. For the

selection of multiple analysis tools, enterprises

should choose appropriate analysis tools and

technologies according to the project and market clear

Problems and Solutions of Data Analytics in Business

125

needs and goals. In addition, a variety of analysis

tools and techniques can be selected to get different

results. Multiple small-scale pilot projects can be

established based on the different results of the

analysis before applying them to the whole project, so

as to test out the most accurate and compliant analysis

tools. For professionals, companies should increase

the education and training of analysts to improve the

data analysis ability of employees. In addition, it can

also be improved by external factors, such as the

introduction of excellent data analysis talents with

high technical literacy, and the cooperation with

professional data analysis organizations to complete

the project. By further improving the tools and

talents, the accuracy of data analysis will be greatly

guaranteed. In addition, the accuracy can be ensured

by adjusting the model itself. Enterprises can also

continuously optimize the analysis model and assess

the accuracy and applicability of the model through

the verification of historical data (Vanwinckelen,

2012). For example, previous historical data can be

used as a validation set to observe the fitting effect of

the analytical model, so as to adjust and analyze the

accuracy of the model. The accuracy of the prediction

can also be improved by using several different

models for prediction and analyzing the results

comparatively. In conclusion. The prediction of the

accuracy of data analysis must not be too high, when

the enterprise is making decisions, it is necessary to

consider the uncertainty of the prediction results and

reduce the risk of decision-making by developing a

variety of response options and risk management

strategies.

3.4 Enhancing Collaboration Across

Sectors

Firstly, this paper must focus on strengthening the

synergy between all parts of the organization. This

means strengthening the communication and

collaboration between the data analysis team and the

business units to ensure that the analysis results are

highly aligned with the actual business needs. In order

to achieve this goal, this paper should establish a “

double-check ” working concept, i.e., in every

project, the business department and the data analysis

department should communicate and report to each

other twice. This minimizes errors and ensures that

both parties have the same understanding and

expectations of the project. Through frequent

communication and feedback, this paper is able to

identify and resolve potential problems in a timely

manner, thereby improving the quality and efficiency

of the project. Second, to avoid errors due to poor

presentation and understanding, this paper should

present data and results in a more visual way. This

can be done through the use of charts, documents and

other forms of presenting complex data processing

processes and results so that other departments can

understand and apply the information in a more

intuitive way. The impact of visualization in big data

analysis is very important for business. It is a

technique of displaying information in a graphical

format that makes it easy for users to understand the

information (Mahajan, & Gokhale, 2017). For

example, this paper can create data dashboards,

reports, and visual models that present key metrics

and trends to business units in a clear and

understandable way. As a result, task assignment and

execution will become more efficient and precise

because everyone will be able to make decisions

based on the same understanding and data. Finally,

before each project acceptance, this paper should

convene a meeting of all project department heads to

provide joint feedback on the results and their

application in decision-making and action, and to

understand the specific progress so far. During the

meeting, departments can share their experiences and

findings, make suggestions for improvement, and

adjust data analysis and optimization strategies based

on feedback from other departments. This cross-

departmental cooperation and communication will

help ensure smooth progress and provide valuable

lessons learned for future projects. By working

together, this paper can continue to improve the

quality and value of data analysis and provide strong

support for business development. In conclusion,

through enhanced collaboration, visual data

presentation and cross-departmental feedback

mechanisms, this paper can improve the accuracy and

usefulness of data analysis to better support business

decisions and actions. This will help enterprises gain

advantages in the competitive market and realize

sustainable development.

4 CONCLUSIONS

In the current business environment, data analytics

has undoubtedly become a core tool for enterprises to

gain competitive advantage. However, data analytics

faces many challenges in its practical application.

Through in-depth analysis of four major problems:

data quality, data leakage, inaccurate data analysis,

and data analysis application, this study proposes

targeted solutions, including the establishment of a

sound data processing mechanism, the enhancement

of data encryption technology, the improvement of

ECAI 2024 - International Conference on E-commerce and Artiﬁcial Intelligence

126

the match between analytical tools and talents, and

the promotion of cross-departmental collaboration.

These countermeasures can help enterprises deal with

various problems in data analysis more effectively

and ensure the accuracy and security of data, so that

they can develop better with the support of data

analysis.

Future research directions can focus on the

innovation and practical application of data analysis.

Firstly, with the continuous development of artificial

intelligence and machine learning technology, the

technology of data analysis is also improving, how to

better apply these technologies in data analysis to

improve the accuracy of prediction and rationality of

decision-making will be one of the important future

research directions. Secondly, better application of

data analytics to company development and decision

making, more accurate as well as rational application

in business is also a very important research direction

in the future.

REFERENCES

Cheng, L., Liu, F., & Yao, D. (2017). Enterprise data

breach: causes, challenges, prevention, and future

directions. Wiley Interdisciplinary Reviews: Data

Mining and Knowledge Discovery, 7(5), e1211.

Choughri, R., Kamaleddine, K., Soubjaki, M., & Baytieh,

M. (2018). The Challenge of Data Accuracy in Business

Analytics that Affect Managers Decision Making–Case

Study of Saudi Arabia & Lebanon. IOSR Journal of

Business and Management (IOSR-JBM).

Chu, X., Ilyas, I. F., Krishnan, S., & Wang, J. (2016, June).

Data cleaning: Overview and emerging challenges. In

Proceedings of the 2016 international conference on

management of data (pp. 2201-2206).

Divyashree, S, Christy Jackson Joshua, Abdul Quadir Md,

et al. (2024). Enabling business sustainability for stock

market data using machine learning and deep learning

approaches. Annals of Operations Research

(prepublish), 1-36.

Eltayieb, N., Elhabob, R., Abdelgader, A. M., Liao, Y., Li,

F., & Zhou, S. (2024). Certificateless Proxy Re-

encryption with Cryptographic Reverse Firewalls for

Secure Cloud Data Sharing. Future Generation

Computer Systems.

Hu, Y. (2018). Marketing and business analysis in the era

of big data. American Journal of Industrial and

Business Management, 8(07), 1747.

Huang, C., Jiang, W., Wu, J., & Wang, G. (2020).

Personalized review recommendation based on users’

aspect sentiment. ACM Transactions on Internet

Technology (TOIT), 20(4), 1-26.

Mahajan, K., & Gokhale, L. A. (2017). Significance of

digital data visualization tools in big data analysis for

business decisions. International Journal of Computer

Applications, 975, 8887.

Vanwinckelen, G., Blockeel, H., De Baets, B., Manderick,

B., Rademaker, M., & Waegeman, W. (2012). On

estimating model accuracy with repeated cross-

validation. In BeneLearn 2012: Proceedings of the 21st

Belgian-Dutch Conference on Machine Learning (pp.

39–44).

Wang, Yitong. (2024). An Introduction to the Application

of Big Data Analytics Technology in Business

Intelligence. Engineering and Management Science (3).

Problems and Solutions of Data Analytics in Business

127