Data, Analytics, Data Science, and Artificial Intelligence.
Data is a very crucial resource to organizations as it contains valuable insights that inform an organization about the nature of the internal and external environment. Today, data enables organizations to make decisions on business operations that guide their operation systems, service delivery processes, and management. The data originating from various equipment and devices like sensors, actuators, and cameras in companies would also provide inputs for smart systems to understand the nature of the environment and initiate action through artificial intelligence and respond effectively. Data analytics help organizations to extract essential details from data and develop strategic plans that contribute to the particular company’s success in the harsh market conditions (Carlos, Kahn, & Halabi, 2018).
In the study of the essence of data in development, it is crucial to identify the following aspects in the process of extracting knowledge from data.
- Importance of data in analytics.
Data are the main ingredient for any BI, data science, and business analytics initiative and they can be viewed as the raw material for what popular decision technologies produce—information, insight, and knowledge (RAMESH et al., 2020). Data is the idea behind the formulation of modern analytics as organizations seek to extract value from unordered data and find algorithms to create visualizations, correlations, patterns, and trends out of data.
- The main inputs and outputs to the analytics continuum considering the new and broad definition of business intelligence.
Data analysis is essential in improving the nature of operation of an organization by developing insights that enable the organization to learn about the customers, the competitors and the business processes. The data inputs to the analytics continuum include the data from these sectors (the customer base, the competitors and business process). The input data consists of the texts, images, voices, web content, comments, and figures that serve as the input content from various sources. The output data would be in the form of patterns, trends, correlations, and knowledge that guide the users and the management in decision making.
- The source and nature of the data for business analytics.
The business processes that provide data for the analytics continuum include the ERP (enterprise resource planning), CRM (customer relationship management), and SCM (supply chain management). The customers and competitors would provide information over the internet like Instagram, Facebook, linked in, twitter, Snapchat, or YouTube. The machines that deploy the internet of things to gather data from the embedded elements and devices also generate input data. The nature of the data would be structured, semi-structured, or unstructured. Unstructured and semi-structured data involves a combination of textual, imagery, voice and web content. Data mining algorithms usually use structured data.
- What are the most common metrics that make for analytics-ready data?
Making data analytics-ready for prediction means that data sets must be transformed into a flat-file format and made available for ingestion into those predictive algorithms (RAMESH et al., 2020). The metrics that define the readiness of data for analytics include:
- Data source reliability. This description defines the originality and appropriateness of the storage medium where the data are obtained (RAMESH et al., 2020). The collection of data from the source eliminates the possibility of data misinterpretation or data transformation as data moves from the source to another destination.
- Data content accuracy. Data accuracy defines the level of correctness that would provide a good match of the required data for the analytics problem.
- Data accessibility. The data needs to be obtained easily; the location for storage, the storage medium would force the access process to include transformations and mergers to gather reliable data.
- Data security and data privacy. Protection of data ensures that only those who are allowed to access data can reach it.
- Data richness. The data should be comprehensive enough and contain all the required elements and is complete.
- Data consistency. The data from various locations need to be merged correctly to ensure that the data for the same subject is integrated and combined successfully. Other aspects that measure the readiness of data include data currency/timeliness, data granularity, data validity, and data relevancy.
Exercise 12.
Go to data.gov—a U.S. government-sponsored data portal that has a considerable number of data sets on a wide variety of topics ranging from healthcare to education, climate to public safety. Pick an item that you are most passionate about. Go through the topic-specific information and explanation provided on the site. Explore the possibilities of downloading the data, and use your favourite data visualization tool to create your meaningful information and visualizations.
Chapter 4. Data Mining.
Data mining is one of the processes of extracting information from data through transformations that form trends, patterns, and correlations in data. It is essential to provide answers to the following questions in the study of data mining.
- Define data mining. Why are there many names and definitions for data mining?
Data mining is a process for developing intelligence in the form of actionable information or knowledge from collected, organized, and stored data through discovering unknown patterns in data. Data mining has become a common practice for organizations as they learn more about their customers as part of analytical decision making. The definition of data mining is stretching to include other forms of data analysis as the user seek to increase their sales by popularizing the data mining label.
- What are the main reasons for the recent popularity of data mining?
The popularity of data mining develops from:
- Increased intensity of competition at a global scale because of rapidly-varying demands of the increasingly saturated marketplace.
- Increased recognition of the value existing in abundant data sources.
- More combined, integrated, or consolidated database records that enable a single view of the attributes of customers, vendors, and transactions.
- Existence of data warehouses that consolidate databases and other repositories in a single location.
- The increase in data processing and storage is drastically increasing.
- Increased efficiency in data storage and processing through reduction of the cost of hardware and software.
- Most organizations and businesses are adopting the conversion of information resources to non-physical forms.
- Discuss what an organization should consider before deciding to purchase data mining software.
An organization should consider the condition of the market, the number and effectiveness competitors and the rate of generation of changes in the market behaviour. They would, therefore, realize that they would need to meet the demands of the market and have a competitive advantage as they use data mining software that would provide reliable predictive accuracy, speed, robustness, scalability, and interpretability.
- Distinguish data mining from other analytical tools and techniques.
Data mining is a process of knowledge extraction that use techniques and tools that are rooted in traditional analytical analysis and artificial intelligence work. Still, it is getting extensive recognition in its ability to solve diverse problems like fraud detection and understanding markets, competitors and the business processes. Data mining tools can combine spreadsheets and other software development tools to enable quick and easy analyzation of mined data.
- Discuss the leading data mining methods. What are the fundamental differences among them?
Data mining methods are grouped as either prediction data mining, association data mining, and segmentation data mining.
Prediction methods involve classification, regression and time-series processes. Prediction methods use trends and patterns from data to forecast on the probability of experiencing an event in future.
Association mechanism includes market-basket analysis, link analysis, and sequence analysis. Association method enables researchers to discover new relationships in large databases.
The segmentation methods include clustering and outlier analysis. Clusters identify natural groupings of things based on their known characteristics, such as assigning customers in different segments based on their demographics and past purchase behaviours (RAMESH et al., 2020).
Exercise 1.
Visit teradatauniversitynetwork.com. Identify case studies and white papers about data mining. Describe recent developments in the field of data mining and predictive modelling.
References.
Carlos, R. C., Kahn, C. E., & Halabi, S. (2018). Data science: big data, machine learning, and
artificial intelligence. Journal of the American College of Radiology, 15(3), 497-498.
RAMESH. DELEN SHARDA (DURSUN. TURBAN, EFRAIM.). (2020). ANALYTICS, DATA
SCIENCE, & ARTIFICIAL INTELLIGENCE: Systems for Decision Support, Global Edition. PEARSON EDUCATION Limited.