AI and Predictive Analytics

Velibor Božić
General hospital Koprivnica
veliborbozic@gmail.com
INTRODUCTION
Predictive analytics (1) is the use of data, statistical algorithms, and machine learning techniques to
analyze historical data and make predictions about future events or outcomes. It involves the use of advanced
analytics techniques to identify patterns, trends, and relationships in data that can be used to make informed
predictions or forecasts. Predictive analytics can be applied to a wide range of fields and industries, including
business, finance, marketing, healthcare, sports, and many others.
Predictive analytics typically involves several key steps, including data collection and preparation, data
analysis, model development, model validation, and model deployment. Data is collected from various sources,
cleaned and pre-processed to remove inconsistencies and errors, and then analyzed to identify relevant
patterns and trends. Statistical algorithms and machine learning techniques are used to develop predictive
models based on the historical data, which can then be validated using different techniques to assess their
accuracy and reliability. Once validated, these models can be deployed in real-world scenarios to make
predictions about future events or outcomes.
Predictive analytics can provide valuable insights and help organizations make data-driven decisions,
optimize operations, identify opportunities, and mitigate risks. It has applications in various fields, such as
predicting customer behaviour, identifying potential fraud or credit risk, optimizing marketing campaigns,
predicting disease outbreaks, forecasting demand for products, predicting equipment failures, and many
others. However, it is important to carefully consider ethical implications, address potential biases, ensure data
privacy, and use predictive analytics as a tool to support human decision-making.
AI (Artificial Intelligence) and predictive analytics (2) are two closely related fields. AI involves the
development of computer programs and systems that can perform tasks that normally require human
intelligence, such as understanding natural language, recognizing patterns in data, and making decisions based
on data analysis. Predictive analytics, on the other hand, is the use of statistical algorithms and machine
learning techniques to analyze data and make predictions about future events or trends.
AI and predictive analytics are often used together to create more accurate and efficient predictions.
By using AI algorithms to analyze data, businesses can identify patterns and relationships that may be difficult
for humans to spot. This can help them make more accurate predictions about customer behaviour, market
trends, and other factors that can affect their business.
Some common applications of AI and predictive analytics include fraud detection, credit risk analysis,
marketing campaign optimization, and supply chain management. In each of these areas, businesses use AI and
predictive analytics to analyze large volumes of data and make predictions about future events. This can help
them make better decisions, reduce costs, and improve their overall performance.
AI and predictive analytics are powerful tools for businesses looking to gain a competitive edge in
today's data-driven economy. By leveraging these technologies, businesses can gain insights that would
otherwise be impossible to obtain and make more accurate predictions about future events.
TOOLS FOR PREDICTIVE ANALYTICS
There are several tools available for predictive analytics that businesses can use to analyze their data
and make predictions about future events. Here are some of the most popular tools (3):
Python. Python is a popular programming language used for data analysis and machine learning. It offers a
wide range of libraries and tools for predictive analytics, including scikit-learn, TensorFlow, and PyTorch.
R. R is a programming language and software environment for statistical computing and graphics. It is widely
used for predictive analytics and has a large community of users and developers.
SAS. SAS is a software suite for advanced analytics, business intelligence, and data management. It offers a
range of tools for predictive modelling, data mining, and machine learning
IBM SPSS. IBM SPSS is a software suite for statistical analysis and data mining. It offers a range of tools for
predictive analytics, including decision trees, neural networks, and regression analysis.
Microsoft Azure Machine Learning. Microsoft Azure Machine Learning is a cloud-based service for building and
deploying machine learning models. It offers a range of tools for data preparation, model training, and
deployment.
Google Cloud AI Platform. Google Cloud AI Platform is a cloud-based service for building and deploying machine
learning models. It offers a range of tools for data preparation, model training, and deployment.
KNIME. KNIME is an open-source data analytics platform that allows users to build data pipelines and
workflows. It offers a range of tools for data mining, machine learning, and predictive analytics.
These are just a few of the many tools available for predictive analytics. When choosing a tool,
businesses should consider their specific needs and requirements, as well as the complexity of their data and
analysis.
PREDICTIVE ANALYTICS IN DESIGN OF USER INTERFACE
Predictive analytics can also be used in the design of user interfaces to improve the user experience
and optimize user interactions. Here are some ways in which predictive analytics can be applied in the design
of user interfaces (4):
Personalization. Predictive analytics can analyze user data, such as past behaviours, preferences, and usage
patterns, to create personalized user interfaces. This can include personalized content, recommendations, and
suggestions tailored to individual users, making the interface more relevant and engaging.
Context-awareness. Predictive analytics can analyze contextual information, such as location, device type, and
time of day, to dynamically adapt the user interface to the user's context. For example, a weather app may
change its interface based on the user's current location and weather conditions, or an e-commerce app may
show different products based on the user's browsing history.
Predictive search. Predictive analytics can enhance search functionality by predicting and suggesting search
queries or results as the user types. This can save time and effort for users and help them find relevant
information more quickly.
Behaviour prediction. Predictive analytics can analyze user behaviour within an interface to predict their next
actions or intentions. For example, a smart home interface may predict when a user is likely to turn on the
lights based on their historical usage patterns and automatically suggest lighting options accordingly.
Error prevention. Predictive analytics can analyze user interactions to detect potential errors or mistakes in
real-time and provide proactive feedback or suggestions to prevent them. For example, a form filling interface
may use predictive analytics to detect potential mistakes in form fields, such as missing or invalid data, and
provide real-time feedback to the user.
Accessibility. Predictive analytics can be used to analyze user interactions and predict accessibility needs for
users with disabilities. This can include predicting the need for larger font sizes, contrasting colours, or
alternative input methods based on user behaviour and preferences.
User behaviour analysis. Predictive analytics can analyze user behaviour patterns, such as usage frequency,
duration, and engagement, to understand user preferences and optimize the user interface accordingly. This
can help designers identify pain points, areas for improvement, and opportunities for innovation in the user
interface design.
By using predictive analytics in the design of user interfaces, designers can create more personalized,
context-aware, and efficient interfaces that enhance the user experience and optimize user interactions.
However, it is important to consider ethical considerations, such as data privacy and transparency, and ensure
that predictive analytics is used as a tool to support user needs and preferences, rather than intrude on their
privacy or manipulate their behaviours.
THE PROCCESS OF PREDICTIVE ANALITICS
The process of predictive analytics involves several steps that businesses can follow to analyze their
data and make predictions about future events. Here are the typical steps involved in the process (5):
Define the problem. The first step in predictive analytics is to define the problem that you want to solve. This
involves identifying the business question or decision that you want to make and determining the data that you
need to collect and analyze.
Collect and prepare the data. The next step is to collect the data that you need and prepare it for analysis. This
may involve cleaning and transforming the data to ensure that it is accurate and complete.
Explore the data. Once you have prepared the data, you can explore it to identify patterns, relationships, and
trends. This may involve using descriptive statistics, visualization tools, and other techniques to gain insights
into the data.
Build a predictive model. Once you have explored the data, you can build a predictive model using statistical
algorithms, machine learning techniques, or other methods. This involves selecting the appropriate model and
training it on the data to make predictions about future events.
Evaluate the model. After building the predictive model, you should evaluate its performance to determine
how well it is able to make predictions. This involves testing the model on a separate set of data and measuring
its accuracy and precision.
Deploy the model. Once you have evaluated the model, you can deploy it in your business to make predictions
about future events. This may involve integrating the model into your business processes and systems, or using
it to make strategic decisions.
Monitor and refine the model. Finally, it is important to monitor the performance of the predictive model over
time and refine it as needed. This may involve updating the model with new data or adjusting the model
parameters to improve its accuracy and performance.
The process of predictive analytics involves several steps that businesses can follow to analyze their
data and make predictions about future events. By following these steps, businesses can gain insights into their
data and make better decisions that can help them achieve their goals.
THE WAYS HOW BUSINESS HELPING AI IN THE PROCCESS OF PREDICTIVE
ANALITICS
There are several ways that businesses can help AI in the process of predictive analytics to improve the
accuracy and effectiveness of their models. Here are some of the key ways to assist AI (6):
Define the problem clearly. One of the most important ways to help AI in predictive analytics is to define the
problem clearly. This involves identifying the business question or decision that you want to make and
determining the data that you need to collect and analyze. By providing clear and detailed information about
the problem, you can help AI to better understand the goals and objectives of the project.
Gather high-quality data. Another important way to help AI is to gather high-quality data that is relevant to the
problem you are trying to solve. This may involve collecting data from a variety of sources and ensuring that it
is accurate, complete, and up-to-date. By providing high-quality data, you can help AI to make more accurate
predictions and reduce the risk of errors or biases.
Use appropriate algorithms and models. The choice of algorithms and models can have a significant impact on
the accuracy and effectiveness of predictive analytics. It is important to select appropriate algorithms and
models that are suitable for the problem at hand and the type of data being analyzed. By choosing the right
algorithms and models, you can help AI to make more accurate predictions and avoid overfitting or underfitting
the data.
Monitor and refine the model. Even the most accurate and effective predictive models can become less
accurate over time due to changes in the data or other factors. It is important to monitor the performance of
the model over time and refine it as needed to ensure that it continues to provide accurate and useful
predictions. This may involve updating the model with new data, adjusting the model parameters, or
incorporating new features into the model.
 There are several ways that businesses can help AI in the process of predictive analytics to improve the
accuracy and effectiveness of their models. By providing clear and detailed information about the problem,
gathering high-quality data, using appropriate algorithms and models, and monitoring and refining the model
over time, businesses can achieve more accurate and effective predictions that can help them achieve their
goals.
USING AI IN PREDICTING ANALYSIS
Using AI in predictive analysis involves leveraging advanced algorithms and machine learning
techniques to analyze historical data and make predictions about future events or outcomes. AI can be used in
various ways in predictive analysis, depending on the specific domain or application. Here are some examples
of how AI can be used in predictive analysis (7), (8):
Fraud Detection. AI can analyze large amounts of data, such as transaction records, user behaviour, and
historical patterns, to detect potential fraud or anomalous activities in real-time. Machine learning algorithms
can learn from past fraud cases and identify patterns or anomalies that may indicate fraudulent activities,
helping businesses or organizations prevent financial losses.
Credit Risk Analysis. AI can analyze diverse data sources, such as credit history, income, employment status,
and other relevant variables, to assess credit risk and predict the likelihood of default for loan applicants.
Machine learning models can be trained on historical data to identify patterns and assess credit risk, assisting
financial institutions in making informed lending decisions.
Marketing Campaign Optimization. AI can analyze vast amounts of customer data, such as demographic
information, purchase history, online behaviour, and social media interactions, to optimize marketing
campaigns. Machine learning algorithms can segment customers, identify preferences, and predict responses
to different marketing strategies, helping businesses target their marketing efforts more effectively.
Supply Chain Management. AI can analyze various factors, such as demand, transportation costs, inventory
levels, and production capacity, to optimize supply chain operations. Machine learning algorithms can make
predictions about demand patterns, optimize inventory levels, and optimize transportation routes, helping
businesses streamline their supply chain processes and reduce costs.
Healthcare. AI can analyze patient data, such as electronic health records, sensor data, and other relevant
variables, to predict disease outcomes, identify potential health risks, and optimize treatment plans. Machine
learning models can analyze large datasets and identify patterns that may not be apparent to human clinicians,
assisting in early diagnosis, personalized treatment planning, and disease management.
Human Resources. AI can analyze employee data, such as performance metrics, engagement surveys, and
feedback, to predict employee turnover, identify potential talent, and optimize workforce planning. Machine
learning algorithms can identify patterns of employee behavior, predict potential attrition, and help businesses
take proactive measures to retain top talent.
Energy Management. AI can analyze data from sensors, meters, weather forecasts, and other sources to
optimize energy usage, predict equipment failures, and optimize maintenance schedules. Machine learning
models can identify patterns and trends in energy usage, predict potential issues, and optimize energy
management strategies for improved efficiency and cost savings.
These are just a few examples of how AI can be used in predictive analysis across various domains and
industries. The use of AI in predictive analysis can provide valuable insights, optimize decision-making, and
improve operational efficiency in diverse areas of application. However, it is important to consider ethical
considerations, such as data privacy, transparency, and fairness, and ensure that the use of AI in predictive
analysis is aligned with ethical guidelines and regulations.
AI can be used in a variety of ways to enhance predictive analytics in various business domains, such as
fraud detection, credit risk analysis, marketing campaign optimization, and supply chain management. Here are
some examples:
Fraud Detection. AI algorithms can analyze large amounts of data to detect patterns and anomalies that
indicate fraudulent activity. For example, banks can use AI-powered systems to analyze customer transactions
and detect unusual patterns of behaviour, such as sudden large transfers or purchases made in unusual
locations. Similarly, insurance companies can use AI to analyze claims data and identify fraudulent claims.
Credit Risk Analysis. AI algorithms can analyze customer data, credit history, and other relevant factors to
predict credit risk and make more accurate lending decisions. For example, banks and other financial
institutions can use AI-powered systems to analyze credit scores, income data, and other relevant factors to
determine the creditworthiness of a borrower and set appropriate interest rates.
Marketing Campaign Optimization. AI can be used to analyze customer data and identify patterns that indicate
which marketing campaigns are most effective. This can help companies to optimize their marketing campaigns
and target the right customers with the right message. For example, a retailer can use AI to analyze customer
purchase history and other data to create personalized marketing messages that are more likely to resonate
with individual customers.
Supply Chain Management. AI can be used to optimize supply chain operations and improve efficiency by
predicting demand, identifying potential disruptions, and optimizing inventory management. For example, a
manufacturer can use AI to analyze historical sales data and predict future demand for their products, enabling
them to optimize their production schedule and ensure that they have enough inventory to meet customer
needs.
AI can be a powerful tool for enhancing predictive analytics in a variety of business domains, including
fraud detection, credit risk analysis, marketing campaign optimization, and supply chain management. By
analyzing large amounts of data and identifying patterns and trends, AI-powered systems can help businesses
to make more informed decisions and improve their overall performance.
USING AI IN PREDICTIVE ANALYTICS OF PATIENT BEHAVIOUR IN HOSPITAL
AI can be used in predictive analytics to better understand and predict patient behaviour in hospitals.
By analyzing large amounts of data from various sources, including electronic health records, medical devices,
and patient feedback, AI-powered systems can help healthcare providers to identify patterns and trends that
indicate patient behaviour and predict future behaviour.
Here are some examples of how AI can be used in predictive analytics of patient behaviour in hospitals
(9),(10):
Patient Flow Management. AI can be used to predict patient admission and discharge times, which can help
hospitals optimize their bed utilization and reduce wait times. For example, AI can analyze historical patient
admission and discharge data, as well as real-time patient data, to predict how long patients are likely to stay in
the hospital and when they are likely to be discharged.
Patient Monitoring. AI can be used to monitor patient behaviour in real-time, alerting healthcare providers to
potential issues before they become serious. For example, AI can analyze patient data from medical devices to
detect changes in vital signs or other indicators of potential health problems.
Patient Satisfaction. AI can be used to analyze patient feedback and identify trends that indicate satisfaction or
dissatisfaction with the care provided. For example, AI can analyze patient feedback from surveys, social media,
and other sources to identify common themes and issues that may be affecting patient satisfaction.
Patient Engagement. AI can be used to predict patient engagement with their care plan, helping healthcare
providers to identify patients who may be at risk of non-compliance and intervene early. For example, AI can
analyze patient data, including medication adherence, appointment attendance, and other factors, to predict
which patients are most likely to follow their care plan and which may need additional support.
AI can be a powerful tool for predicting patient behaviour in hospitals, helping healthcare providers to
optimize patient flow, monitor patient health, improve patient satisfaction, and engage patients in their care.
By leveraging the technology of AI, healthcare providers can deliver more personalized, effective care to their
patients.
ROLES OF AI AND PREDICTIVE ANALYTICS IN DESIGN USER INTERFACE
ESPECIALLY FOR VULNERABLE GROUPS OF USERS
AI and predictive analytics can play important roles in designing user interfaces for vulnerable groups
of users. Here are some of the key ways that AI and predictive analytics can be used (11):
Personalization. AI can be used to personalize the user interface based on the individual needs and preferences
of the user. For example, predictive analytics can be used to predict the user's needs based on their past
behaviour and provide customized recommendations or suggestions. This can be especially useful for
vulnerable users who may have difficulty navigating complex user interfaces.
Accessibility. AI can also be used to improve accessibility for vulnerable users. For example, predictive analytics
can be used to identify users who may have difficulty seeing or hearing certain content and provide alternative
formats that are easier to access. Additionally, AI can be used to identify and correct accessibility issues in real-
time, such as providing audio descriptions or closed captions for videos.
Usability testing. AI can also be used to conduct usability testing with vulnerable users. For example, predictive
analytics can be used to identify users who may have difficulty using certain features or functions of the user
interface and provide feedback on how to improve the design. This can help designers to identify and address
potential issues before the product is released to the public.
Emotional engagement. AI can also be used to create emotionally engaging user interfaces that are more
appealing and engaging for vulnerable users. For example, predictive analytics can be used to identify
emotional triggers that are most effective for different users and tailor the user interface to their emotional
needs. This can help to create a more positive user experience and increase user engagement.
AI and predictive analytics can play important roles in designing user interfaces for vulnerable groups
of users. By using AI to personalize the user interface, improve accessibility, conduct usability testing, and
create emotionally engaging experiences, designers can create user interfaces that are more effective,
engaging, and inclusive.
TECHNOLOGIES OF AI IN DESIGNING USER INTERFACE
Artificial Intelligence (AI) can be used in designing user interfaces to improve the user experience by
personalizing the interface, providing better accessibility, and enhancing usability. Here are some of the key
technologies of AI that can be used in designing user interfaces (12):
Natural Language Processing (NLP). NLP technology enables computers to understand human language,
allowing for the creation of interfaces that can respond to natural language commands or questions. This can
be especially useful for vulnerable users who may have difficulty navigating complex user interfaces.
Computer Vision. Computer vision technology enables computers to interpret and analyze visual information,
such as images or videos. This can be used to create interfaces that can recognize and respond to user gestures
or facial expressions, improving usability and engagement.
Machine Learning. Machine learning technology can be used to analyze user data and behavior to provide
personalized recommendations and suggestions. For example, machine learning algorithms can analyze a user's
past behaviour to predict what features or content they are most likely to use or enjoy.
Sentiment Analysis. Sentiment analysis technology enables computers to analyze and understand human
emotions and attitudes, allowing for the creation of interfaces that can respond to users' emotional needs. For
example, sentiment analysis algorithms can identify when a user is feeling frustrated or confused and provide
customized assistance or guidance.
Natural Language Generation (NLG). NLG technology enables computers to generate natural language text,
which can be used to provide users with personalized feedback or recommendations. For example, NLG
algorithms can generate personalized summaries of user activity or provide customized responses to user
inquiries.
The technology of AI can be used to create user interfaces that are more personalized, accessible, and
engaging for vulnerable users. By leveraging NLP, computer vision, machine learning, sentiment analysis, and
NLG technologies, designers can create user interfaces that are more effective, user-friendly, and inclusive.
EXAMPLES OF HOW AI HAS BEEN USED IN DESIGNING USER INTERFACES
 The Seeing AI app (13), (14) developed by Microsoft uses computer vision and NLP technology to help
people with visual impairments navigate the world around them. The app can recognize and describe
objects, read text, and identify people, helping users to interact with their environment more effectively.
 The Sesame Enable app uses computer vision technology to enable people with disabilities to control their
smartphones and other devices using head movements. The app allows users to navigate the interface,
type messages, and access apps using only their head movements.
 The Replika app uses machine learning and NLG technology to create a personalized chatbot that can
provide emotional support to users. The app learns from user input to create a virtual companion that can
provide customized feedback, advice, and encouragement.
 The EmoReact platform developed by Affectiva uses sentiment analysis technology to analyze users'
emotional responses to content, such as videos or advertisements. The platform can be used to create
more emotionally engaging user interfaces that are better suited to the needs and preferences of
individual users.
 The Grammarly writing assistant uses NLP technology to provide personalized feedback on grammar,
spelling, and style. The app analyzes user input to provide customized suggestions and recommendations,
improving the user experience and helping users to communicate more effectively.
These are just a few examples of how AI has been used in designing user interfaces for vulnerable
users. By leveraging the technology of AI, designers can create interfaces that are more personalized,
accessible, and engaging, helping to improve the user experience for everyone.
STUDIES ABOUT USING AI IN PREDICTIVE ANALYSIS
There have been numerous studies and research papers published on the use of AI in predictive
analytics. Here are a few examples:
 "Artificial intelligence and machine learning in predictive analytics for healthcare" by Rajkomar et al. (2018)
- This study explores the use of AI and machine learning in predictive analytics for healthcare, including
predicting patient outcomes and identifying high-risk patients.
 "Predictive Analytics Using Machine Learning Techniques: A Review" by Alzahrani et al. (2020) - This review
paper summarizes recent research on the use of machine learning techniques for predictive analytics in
various domains, including healthcare, finance, and marketing.
 "A review of artificial intelligence and machine learning for automated fraud detection and prediction in
financial and accounting data" by Jabeen et al. (2021) - This paper explores the use of AI and machine
learning for fraud detection and prediction in financial and accounting data, including credit card fraud and
accounting fraud.
 "AI-Driven Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die" by Davenport and
Ronanki (2018) - This book explores the use of AI-driven predictive analytics in various domains, including
marketing, finance, and healthcare, and provides case studies and examples of successful
implementations.
There is a growing body of research and literature on the use of AI in predictive analytics, with
applications in a wide range of domains and industries.
RESULTS OF STUDIES
The results of studies on the use of AI in predictive analytics vary depending on the specific application
and context. However, some common findings and conclusions can be drawn from the literature.
 Improved Accuracy: One of the main advantages of using AI in predictive analytics is improved accuracy
compared to traditional statistical methods. AI algorithms can analyze large amounts of data and identify
complex patterns and relationships that are difficult to detect using traditional methods.
 Increased Efficiency: AI can also improve efficiency by automating the process of data analysis and
prediction, reducing the need for manual labour and increasing the speed of decision-making.
 Enhanced Personalization: AI can also enable more personalized predictions by analyzing individual-level
data and tailoring predictions to specific individuals or groups.
 Ethical Considerations: However, there are also ethical considerations when using AI in predictive
analytics, including concerns about bias, privacy, and the potential for negative consequences for
individuals.
The results of studies on the use of AI in predictive analytics are generally positive, with AI algorithms
showing promise for improving accuracy, efficiency, and personalization. However, it is important to carefully
consider the ethical implications of using AI in predictive analytics and take steps to mitigate potential risks and
negative consequences.
RISKS OF USING AI IN PREDICTIVE ANALYSIS
While AI has the potential to improve predictive analytics in a wide range of applications, there are
also risks associated with its use. Here are some potential risks of using AI in predictive analysis (15):
Bias. AI algorithms may be biased if they are trained on biased data or if the data used to train them is not
representative of the population being analyzed. This can lead to inaccurate predictions or discriminatory
outcomes.
Privacy. Predictive analytics often involve the use of personal data, such as medical records or financial data.
There is a risk that this data could be accessed or used improperly, leading to privacy violations or other
negative consequences.
Overreliance on technology. There is a risk that healthcare providers or other professionals may become overly
reliant on AI predictions and may overlook important information or make poor decisions based on incomplete
or inaccurate data.
Negative consequences for individuals. Predictive analytics may be used to make decisions that have significant
impacts on individuals, such as denying access to credit or medical treatment. There is a risk that these
decisions could be unfair or have negative consequences for individuals.
Lack of transparency. AI algorithms can be complex and difficult to understand, which can make it difficult to
identify and correct errors or biases in the predictions they generate.
It is important to carefully consider the risks and potential negative consequences of using AI in
predictive analysis, and to take steps to mitigate these risks through careful data selection and evaluation,
appropriate safeguards for data privacy, and ongoing monitoring and evaluation of AI predictions.
MITIGATION OF THE RISKS
Here are some ways to mitigate the risks associated with using AI in predictive analysis (16):
Address bias. To mitigate the risk of bias in AI predictions, it is important to ensure that the data used to train
the algorithms is diverse and representative of the population being analyzed. Data should be carefully
selected and evaluated for potential biases, and steps should be taken to address any biases that are identified.
Ensure data privacy. To protect against privacy violations, it is important to ensure that personal data is
collected and stored securely, and that appropriate safeguards are in place to prevent unauthorized access or
use of this data. This may involve implementing encryption or other security measures, as well as complying
with relevant data protection regulations.
Use AI as a tool, not a replacement. It is important to remember that AI should be used as a tool to assist
decision-making, rather than as a replacement for human judgment. Professionals should carefully consider AI
predictions in the context of other relevant information and use their own judgment to make final decisions.
Consider ethical implications. Before using AI in predictive analysis, it is important to carefully consider the
ethical implications of its use, including the potential impacts on individuals and the broader societal
implications. This may involve engaging in discussions with stakeholders and consulting with experts to ensure
that decisions are made in a responsible and ethical manner.
Monitor and evaluate AI predictions. Finally, it is important to monitor and evaluate AI predictions over time to
ensure that they continue to be accurate and unbiased, and to identify and address any errors or biases that
may arise. This may involve ongoing training of AI algorithms and careful evaluation of their performance in
real-world applications.
FURTHER RESEARCHES ABOUT USING AI IN PREDICTIVE ANALYSIS
There is ongoing research on the use of AI in predictive analysis across a wide range of applications.
Some areas of research include (17):
Healthcare. AI is being studied for its potential to improve patient outcomes and reduce healthcare costs by
predicting patient behaviour, optimizing treatment plans, and improving disease detection and diagnosis.
Finance. AI is being used to predict credit risk and fraud, as well as to optimize investment strategies and
improve financial planning and management.
Marketing. AI is being used to optimize marketing campaigns by predicting consumer behaviour and tailoring
marketing strategies to specific audiences.
Supply Chain Management. AI is being used to predict demand and optimize supply chain operations,
improving efficiency and reducing waste.
Cyber security. AI is being used to predict and prevent cyber attacks, as well as to detect and respond to
security breaches more quickly and effectively.
Climate Science. AI is being used to predict the impacts of climate change and to develop more accurate
climate models, enabling better planning and mitigation efforts.
The use of AI in predictive analysis has the potential to improve accuracy, efficiency, and
personalization in a wide range of applications. Ongoing research will continue to explore new applications and
refine existing techniques, with the goal of maximizing the benefits of AI while minimizing the risks.
CONCLUSION
AI has significant potential to improve predictive analytics in many fields, including healthcare, finance,
marketing, supply chain management, cyber security, and climate science. By using AI to analyze large amounts
of data and make predictions about future outcomes, professionals can make more informed decisions and
optimize their operations for better results. However, the use of AI in predictive analysis also carries risks,
including the potential for bias, privacy violations, overreliance on technology, negative consequences for
individuals, and lack of transparency.
To mitigate these risks, it is important to carefully consider the ethical implications of AI use, monitor
and evaluate AI predictions over time, and take steps to ensure data privacy and diversity. Ongoing research
will continue to explore new applications for AI in predictive analytics and refine existing techniques, with the
goal of maximizing the benefits of this technology while minimizing the risks.
REFERENCES
1. Eckerson, W. W. (2007). Predictive analytics. Extending the Value of Your Data Warehousing
Investment. TDWI Best Practices Report, 1, 1-36.
2. Letourneau-Guillon, L., Camirand, D., Guilbert, F., & Forghani, R. (2020). Artificial intelligence
applications for workflow, process optimization and predictive analytics. Neuroimaging Clinics, 30(4),
e1-e15.
3. Yun, C., Shun, M., Junta, U., & Browndi, I. (2022). Predictive Analytics: A Survey, Trends, Applications,
Opportunities’ and Challenges for Smart City Planning. International Journal of Computer Science and
Information Technology, 23(56), 226-231.
4. H. Khalajzadeh, M. Abdelrazek, J. Grundy, J. Hosking and Q. He, "A Survey of Current End-User Data
Analytics Tool Support," 2018 IEEE International Congress on Big Data (BigData Congress), San
Francisco, CA, USA, 2018, pp. 41-48, doi: 10.1109/BigDataCongress.2018.00013.
5. Nyce, C., & Cpcu, A. (2007). Predictive analytics white paper. American Institute for CPCU. Insurance
Institute of America, 9-10.
6. Kumar, M., Shenbagaraman, V. M., Shaw, R. N., & Ghosh, A. (2021). Predictive data analysis for energy
management of a smart factory leading to sustainability. In Innovations in Electrical and Electronic
Engineering: Proceedings of ICEEE 2020 (pp. 765-773). Springer Singapore.
7. Armstrong, S., & Sotala, K. (2015). How we’re predicting AI–or failing to. Beyond Artificial Intelligence:
The Disappearing Human-Machine Divide, 11-29.
8. Gomathi, S., Kohli, R., Soni, M., Dhiman, G., & Nair, R. (2020). Pattern analysis: predicting COVID-19
pandemic in India using AutoML. World Journal of Engineering, 19(1), 21-28.
9. Davenport, T., & Kalakota, R. (2019). The potential for artificial intelligence in healthcare. Future
healthcare journal, 6(2), 94.
10. Chen, M., & Decary, M. (2020, January). Artificial intelligence in healthcare: An essential guide for
health leaders. In Healthcare management forum (Vol. 33, No. 1, pp. 10-18). Sage CA: Los Angeles, CA:
SAGE Publications.
11. Asan, O., Bayrak, A. E., & Choudhury, A. (2020). Artificial intelligence and human trust in healthcare:
focus on clinicians. Journal of medical Internet research, 22(6), e15154.
12. Ferreira, J. J., & Monteiro, M. S. (2020). What are people doing about XAI user experience? A survey
on AI explainability research and practice. In Design, User Experience, and Usability. Design for
Contemporary Interactive Environments: 9th International Conference, DUXU 2020, Held as Part of the
22nd HCI International Conference, HCII 2020, Copenhagen, Denmark, July 19–24, 2020, Proceedings,
Part II 22 (pp. 56-73). Springer International Publishing.
13. Johnson, J., & Finn, K. (2017). Designing user interfaces for an aging population: Towards universal
design. Morgan Kaufmann.
14. Quinn, L. (2023).The Impact of AI on UX Design: Opportunities and Challenges. Downloaded:
https://uxplanet.org/the-impact-of-ai-on-ux-design-opportunities-and-challenges-a9e466d319ad.
Downloaded: 4-11-2023.
15. Radanliev, P., De Roure, D., Page, K., Nurse, J. R., Mantilla Montalvo, R., Santos, O., ... & Burnap, P.
(2020). Cyber risk at the edge: current and future trends on cyber risk analytics and artificial
intelligence in the industrial internet of things and industry 4.0 supply chains. Cybersecurity, 3(1), 1-21.
16. Cheatham, B., Javanmardian, K., & Samandari, H. (2019). Confronting the risks of artificial
intelligence. McKinsey Quarterly, 2(38), 1-9.
17. Loureiro, S. M. C., Guerreiro, J., & Tussyadiah, I. (2021). Artificial intelligence in business: State of the
art and future research agenda. Journal of business research, 129, 911-926.