Skip to content

8. Future of Big Data

Framework for Pandemic Prediction Using Big Data Analytics 1

  • In Covid-19’s situation, the IoT and big data technologies have played an essential role in fighting against the pandemic. The applications might include:
    • Rapid collection of big data,
    • Visualization of pandemic information,
    • Breakdown of the epidemic risk,
    • Tracking of confirmed cases,
    • Tracking of prevention levels,
    • Adequate assessment of COVID-19 prevention and control.

Descriptive analytics

Attribute Information => Common Statistics => Data Visualization

  • The simplest form of analytics that describes what has happened in the past
  • Detailed information about every attribute present in the data set is provided, including the number of attributes, each attribute’s feature, and the population size of attributes in the data set.
  • Using visual graphs, the summary or detailed description of the raw data is made that is interpretable by humans.
  • In healthcare, it merely identifies the standard statistics of data; for example:
    • the total number of laboratory tests performed,
    • the average age of patients,
    • the number of males and females suffering from particular diseases,
    • the average duration/period of stay in the hospital for patients,
    • the number of people who recovered from particular diseases.
  • Bar or column charts and tables, pie charts, or written narratives are used.

Diagnostic analytics

Data Discovery => Data Mining => Correlation Techniques

In Healthcare: Data Discovery => Symptoms Diagnoses => Symptoms Correlation

  • It is a more advanced form of analytics that helps in understanding why something happened.
  • It helps to examine insight of data that answers the question “Why did it happen?”.
  • It considers different attributes and features information in order to discover the relations.
  • It is likewise known as data discovery, data mining, and correlation techniques.
  • It helps to analyze insight of data and attempts to interpret the causes of events and behaviors.
  • In health care, diagnostic analytics explore the information and make correlations using different attribute information. For example:
    • It may help determine that all of the patient’s symptoms, such as high fever, dry cough, flu, and fatigue, point to the same virus agent.
    • The symptoms and causes behind the diseases are explored
  • Correlation graphs, scatter plots, and heat maps are used.
  • A correlation could be positive if both variables move in the same direction or negative when there is an increase in one variable’s value decrease in other variables’ values.

Predictive analytics

Data Mining => Machine Learning => Deep Learning => Prediction

  • It is a more advanced form of analytics that helps in understanding what is likely to happen in the future.
  • It understands the insights of data and provides possible suggestions to organizations with actionable insights.
  • The recorded data is fed into a machine or deep learning model that considers the data’s key patterns and trends. The model is then applied to current data for prediction.
  • In healthcare, predictive analysis is used for:
    • forecasting disease spread rate,
    • and the chances of patient survival.

Prescriptive analytics

Predictive Data Results => Prescribe/Determine => Possible Actions

In Healthcare: Predictive Data Results => Prescribe/Determine => Expert Opinions => Possible Actions

  • It is the most advanced form of analytics that helps in understanding what actions to take in the future.
  • It takes advantage of predictive data results and facilitates users to “prescribe/determine” various possible actions to implement and direct them toward a solution.
  • It tries to evaluate the effect of future decisions and advise possible outcomes before decision-making.
  • It forecasts what will happen and explains why it will happen, thereby giving suggestions about actions that take benefit from predicted results.
  • It suggests various courses of action and outlines what the potential implications would be for each.
  • In healthcare, it plays an essential role in:
    • the prevention and control of diseases spreading.

Social Big Data: Recent Achievements and New Challenges 2

  • The combination of big data technologies and traditional machine learning algorithms has generated new and interesting challenges in other areas as social media and social networks.
  • These new challenges are focused mainly on problems such as data processing, data storage, data representation, and how data can be used for pattern mining, analyzing user behaviors, and visualizing and tracking data, among others.
  • Social media is a natural source for data analysis; big data is a parallel and massive processing paradigm; and data analysis is a set of algorithms and methods used to extract and analyze knowledge.

❌ ❌ ❌

A Survey on Blockchain for Big Data: Approaches, Opportunities, and Future Directions 3

  • There are many challenges in big data to be tackled for better quality of service, e.g., big data analytics, big data management, and big data privacy and security.
  • Blockchain with its decentralization and security nature has the great potential to improve big data services and applications.
  • The use of blockchain for big data applications in different domains such as smart cities, smart healthcare, smart transportation, and smart grid.
  • Blockchain as a ledger technology has emerged as an attractive solution for providing security and privacy in big data systems. Blockchain can provide high-quality data and secure data sharing for industrial IoT applications.

An Efficient Predictive Analytics System for High-Dimensional Big Data 4

  • An efficient predictive analytics system for high dimensional big data is proposed by enhancing the Scalable Random Forest (SRF) algorithm on the Apache Spark platform.
  • SRF is enhanced by optimizing the hyperparameters and prediction performance is improved by reducing the dimensions.

Overcoming Data Silos Through Big Data Integration 5

A Survey of Predictive Analytics Using Big Data with Data Mining 6

Video Resources 7 8

References

  1. Ahmed, I., Ahmad, M., Jeon, G., & Piccialli, F. (2021). A framework for pandemic prediction using big data analytics. Big Data Research, 25, 100190. https://doi.org/10.1016/j.bdr.2021.100190 ↩

  2. Bello-Orgaz, G., Jung, J. J., & Camacho, D. (2016). Social big data: Recent achievements and new challenges. Information Fusion, 28, 45-59. https://doi.org/10.1016/j.inffus.2015.08.005 ↩

  3. Deepa, N., Pham, Q-V., Nguyen, D. C., Bhattacharya, S., Prabadevi, B., Gadekallu, T. R., Maddikunta, P. K. R., Fang, F., & Pathirana, P. N. (2022). A survey on blockchain for big data: Approaches, opportunities, and future directions. Future Generation Computer Systems, 131, 209-226. https://doi.org/10.1016/j.future.2022.01.017 ↩

  4. Mon Oo, M. C., & Thein, T. (2022). An efficient predictive analytics system for high dimensional big data. Journal of King Saud University - Computer and Information Sciences, 34(1), 1521-1532. https://doi.org/10.1016/j.jksuci.2019.09.001 ↩

  5. Patel, J. (2019). Overcoming data silos through big data integration. International Journal of Education (IJIT), 4 (4). https://famebook.com/jounals/IJIT/paper/IJIT003.pdf ↩

  6. Selvaraj, P., & Marudappa, P. (2018). A survey of predictive analytics using big data with data mining. International Journal of Bioinformatics Research and Applications, 14(3), 269-282. https://www.researchgate.net/profile/Poornima-Selvaraj/publication/326071541_A_survey_of_predictive_analytics_using_big_data_with_data_mining/links/5c51132f92851c22a39a385b/A-survey-of-predictive-analytics-using-big-data-with-data-mining.pdf ↩

  7. FME Channel. (2020, February 11). What is data integration, and how does it work? [Video]. YouTube. https://www.youtube.com/watch?v=zj0ZxjxHOAs ↩

  8. Kelly, K. (2020, July 1). The future of big data. [Video]. YouTube. https://www.youtube.com/watch?v=JBcplC8AFuE ↩