5. Applications of Big Data and Business Analytics

There are various areas of business and industries that have benefited from big data analytics technologies. These areas generate a huge amount of data that require big data analytics process for effective and efficient decision making. These application areas include healthcare, telecommunication, network optimization, travel estimation, retails, financial industries, energy consumption [4,56] to mention but a few. The application areas are explained below while Table 3 outlines the key data sources in these areas and features.

Table 3. Applications and key data sources for big data and business analytics.

Application Key Data Sources Features
Healthcare Electronic health record, patients' information, images, health history data. Support improved health monitoring, study patients' immune systems, activity recommendation for elderly physical health
Financial Industries Financial reports, stock news, blog post, social media, and annual general meeting information Provide a mechanism for fraud detection, mitigate against money laundry and decision making
Network Optimization Network signal information, information between network users, weblog, geo-location data, sensor data, video camera, and network log Efficient network signaling, prediction of network variation, network management and to generate cell deployment information
Travel estimation GPS data, location data, satellite imagery, personal information, call data record(CDR) Provide information for complex route recommendation, location tracking, drone routing for a military operation, emergency situation and infectious disease identification
User Behavior Modeling Log data, social media data, blog post, tweets, and product review Effective and efficient individual service recommendation.
User mobility modeling Location data, GPS Maintain global movement pattern to enable disease containment and transportation planning
Service Recommendation Customer product review, product selection, location data, buying behavior data. Enhanced product buying using customer product review and ascertain weaknesses and strength of products
Energy Consumption Analysis Gas status, consumption pattern data, location data, smart meter reading data, and usage history. Promote green energy, conservation, and efficiency through energy consumption prediction.
Crowdsourcing and sensing Sensing data such as accelerometer, gyroscopes, magnetometer, electrocardiograph (ECG), pulse rate, electromyography (EMG), online questionnaire and survey. Approach for large scale data collection project using a smartphone and online platforms.
Educational development Student information, examination information, student enrollment, course allocation, course contents, Predict student enrollment ratio and dropout rate after particular course or session

  • Healthcare: Improved health is important for economic growth, good physical and mental health. Healthcare industry generates a huge amount of data that can be used to enhance decision making by both doctors and other health practitioners. In addition, the use of big data in healthcare can help to develop a real-time analysis of disease thereby improving the quality of life to the public. There are lots of research in this regards and range from fault tolerance system to support data generation, integration and analysis to continuous monitoring for early detection of an environmental condition that may trigger asthma attack. Moreover, public health care data require big data analytics techniques due to their large scale to track, monitor, store and analyze individual moving objects with their level of exposure to harmful environmental factors in order to ascertain the relationship between the data and environmental risk. Furthermore, big data analytics have played a vital role in predicting the outbreak of diseases such as Ebola virus using call detail records and sensor data to provide feedback mechanism in order to improve quality of healthcare delivery system.
  • Network Optimization: Big data and business analytics approach can be used to design a mobile network to provide efficient services. The area of interest is in content-centric analysis, traffic analysis, network signaling to ensure effective service delivery and quality of service delivery. Network operators can incorporate framework to collect, store and analyze user or core network data for efficient signaling, predict traffic variation, network overload, intelligent network optimization, automatic self-configuration of the network and intelligent transportation development.

  • Travel Estimation: High volume of data generated by mobile users during calls often referred to as call data records (CDRs) has enabled researchers to aggregate, store, process and analyze travel estimation particularly in route recommendation, location tracking, trip generation, commuter origin and destination information and transportation management planning in the developing economy. Mobile big data can also aid route recommendation in a complex environment by deploying smart multimodal platform that utilizes personal information and global constraint. The algorithms monitor the state of the cities in real time and identify the congested route in order to make alternative recommendations. This mechanism is not new as it has seen its applications in drone routing, infectious disease, and hotspot identification and in an emergency situation. To ensure security, the datasets are usually anonymized using computer generated unique identifiers to replace the phone numbers of subscribers. Researches in mobile big data for travel estimate have proven to be important to improve transportation planning.

  • User behavior modeling: User behavior modeling helps to understand navigation patterns in order to develop user-centric applications. These applications are important in anomalies, fraud and spam detection in social media and enable social behavior changes for target marketing.

  • Human mobility modeling: Human beings maintain a regular pattern over a period of time. Consequently, repeating such pattern enables efficient prediction of a global movement and this can be applied in disease containment, transportation planning, emergency situation and prevent the outbreak of diseases by leveraging the social network platform, GPS data, call data record and geo-tagged data through big data analytics methods.

  • Service recommendation: Big data and business analytics approaches have played a vital role in services recommendation, target advertisement using user location information, product review, time and product buying behavior. For instance, a recent study by Salehan and Kim deployed Hadoop and MapReduce to analyze customer review to understand the strengths and weaknesses of the product. This approach helps to determine the predictors of review readership and how to improve sales.

  • Energy consumption analysis: Identification of the amount of energy in the household is a sure way to promote green energy efficiency and conservation. The analysis using big data techniques provides the usage patterns to promote green energy by fitting the electricity supplies with sensors, communication network and analytics engine to digitalize, store and analyze the consumption rate. Moreover, this will help to improve energy sales and return on investment for energy companies.

  • Crowdsourcing and Sensing: Crowdsourcing implemented through opportunistic sensing is an essential source of data for data-driven decision making in a business environment. Many companies employ these techniques to enlist people to perform a specific task for solving complex problems by leveraging smartphone with embedded sensors. Smartphones can be used to source a huge amount of opinion data from the public and then analyze decision-making in an urban emergency, location-based search and similarity services using mobile phone data.

  • Educational development: Educational sector provides rich sources of data for big data analytics processes. These data help to predict learner performances and achievement. Moreover, big data analytics in education play an important role in course content management, personalized recommendation module, development of smart education by leveraging areas such as natural language processing and text summarization. In addition, data generated through massive online courses (MOOCs) helps to identify difficult areas of the subjects and provides support to students in order to enhance teaching and learning.

  • Financial Industries: The adoption of social media and internet-based approaches to financial industries have resulted in the generation of the high volume of data. Therefore, to analyze these data for effective decision-making requires big data techniques. Moreover, analysis of financial statement and data would result in the detection and management of anti-money laundry, financial statement fraud, financial spamming, impersonation, identity theft, and other financial fraud related incidences.

These applications alongside key data sources and features and summarized in Table 3.