A Critical Evaluation of the Big Data Approach to Drug Misuse Data Analytics

Introduction

The purpose of this evaluation is to analyze the application of Big Data analytics to understand in deeper the drug usage pattern and the consequences of drug abuse to develop preventative measures. In order to accelerate progress toward understanding how human health may be altered by drug usage, it is critical to developing powerful analytical methods and visualization tools that can help in capturing the data generated from pharmaceutical records (NIH, 2017).

Big Data analytics is a process of examining a large volume of data  that have hidden patterns, unknown associations, trends in the market and other useful information that can help the organizations to make an informed decision based on information acquired from Big Data analysis. We have different data analytic tools that ensure the storage of data, processing of data, and analyzing and visualizing the data. One of the tools that can be used for data storage is NoSQL. The predictive analytic tool can be used to analyze data based on past events. Hadoop can be used to process data. The analytic tools are an essential part of big data because they have the capability of handling the 4V’s of Big Data (Guru 99, n.d.). They include,

Volume – This is the amount of data produced.

Velocity – The represent the speed in which data are being produced in a given time.

Variety – Data can come up in different formats, for example, images, text, or videos.

Veracity – this is all about the validity of the data.

Word count(246)

Analysis of Big Data analysis tools in approach to drug misuse data analytics.

In big data, data come in large sizes and different structures. Some processes need to be done in big data to analyze the data. To accomplish the analysis of data, we need the necessary tools. We will discuss in-depth the required tools that can be used to perform drug misuse data analytic.

The first step in data analytic is data identifications. Data identifications involve collecting data from different sources, websites, pharmaceutical records, health facilities, or from nation medical centers.  The identified data can be structured, unstructured, or semi-structured. We need to come up with collect to store the data.

Most analytical tools work on top of database frameworks. In our case, we need a database that can hold a large volume of data at a time. NoSQL database can carry a large volume of both structured and unstructured data. They have the capabilities of scaling out. In the dataset provided, some records have more fields than others. NoSQL can handle different data from different areas. Another advantage of using NoSQL is the ability to take data in real-time.

Having information about drug abuse, we can use blockchain to keep the data safe, up-to-date, and to preserve its quality (Sharma, 2019). Blockchain is decentralized, meaning data entered on the blockchain cannot be controlled, or their integrity cannot be altered. Having the data on the blockchain ensures the pharmaceutical can have access to the same data, which is unchangeable (Sharma, 2019).

Another tool we can use in our case is the Hadoop ecosystem. Hadoop ecosystem is made up of different components that can be used together to provide services such as data storage, data analysis, absorption, and maintenance of data (geeks for geek, 2019).

A significant component in the Hadoop ecosystem is HDFS. The Hadoop ecosystem is the tool responsible for storing data. The tool can handle both structured and unstructured data making it appropriate to use for pharmaceuticals. HDFS is made of two components name node and data node. NameNode stores the number of blocks, the locations of the block, and manage the name of the file system (flair, 2019). The data node is where the actual data is stored.

Another component is the Hadoop MapReduce, a framework that writes an application to process a large set of data, both structured and unstructured, stored in HDFS. MapReduce is made up of two functions map() and reduce(). Map() function sorts and filter data and reduce() function is used to perform data aggregation (geeks for geek, 2019).

Another analytic tool is Predictive analytic that is an enabler for Big Data. Predictive analytic uses the past data collected by an organization together with the insight from the business to predict future outcomes. For pharmaceuticals to predict the drug usage pattern, they can use the past drug usage pattern.

Word count (477)

Conclusion

We have used Tableau to analyze the data given by the pharmaceutical. One disadvantage is that Tableau cannot store data. Tableau is expensive tool to purchase and it requires someone to have proper skills to operate. There are security issues with Tableau, and one can easily manipulate confidential information from it.

We can always use big data and blockchain technology to provide a secure way of storing our data. Data is crucial at this time of age. Information is money, and we need to ensure we have to safeguard it from fraudsters or cyber-attack.

We have analyzed different tools for big data analysis. The first tool we analyzed was NoSQL,  a database that has the capability of handling a large volume of data with different structures. The pharmaceuticals obtain data from different drug users. The data collected will always be unstructured because it may not have a fixed pattern, or it may consist of various fields depending on the user. The NoSQL is scalable, meaning new fields can always be added to the database.

The Hadoop ecosystem has the capability of storing data, analyzing the data, and maintain the data, making it is cost-effective and can work on top of NoSQL. Hadoop is made of different components that allow us to store data, filter and sort data, and also provide analysis for the data stored.  In my opinion, I will recommend the use of NoSQL together with the Hadoop framework to provide data analytic for the scenario.

Word count(249)

References

flair, d., 2019. Hadoop Ecosystem and Their Components – A Complete Tutorial. [Online]
Available at: https://data-flair.training/blogs/hadoop-ecosystem-components/[Accessed 15 March 2020].

geeks for geek, 2019. Hadoop ecosystem. [Online]
Available at: https://www.geeksforgeeks.org/hadoop-ecosystem/
[Accessed 14 Match 2020].

Guru 99, n.d. Introduction to BIG DATA. [Online]
Available at: https://www.guru99.com/what-is-big-data.html
[Accessed 14 Match 2020].

NIH, 2017. The Application of Big Data Analytics to Drug Abuse Research. [Online]
Available at: https://grants.nih.gov/grants/guide/pa-files/pa-18-057.html
[Accessed 13 March 2020].

Sharma, A., 2019. How Blockchain and Big Data Complement Each Other. [Online]
Available at: https://hackernoon.com/how-blockchain-and-big-data-complement-each-other-92a1b9f8b38d
[Accessed 14 March 2020].