Abstract

In this era of information, the development of figures has radically expanded, where a colossal amount of data is being delivered from various sources. Because of this huge collection, the worth of information turns into a significant component in each perspective. Data communication is very important to any business - be it small, midsize or Brobdingnagian. Businesses need Data Visualizations to identify data trends at a rapid pace, which would otherwise be tedious. Data Visualization is a robust technology capable of presenting a large dataset in a graphical format. Its centrality goes far beyond merely introducing information to clients. It also assists customers in comprehending information and making decisions. It involves the use of charts, graphs, diagrams, and other info-graphics to communicate data. Surprisingly, a plethora of Data Visualization tools and methodologies have evolved in recent years. This paper conducts an assessment of the most widely used and distributed visualization tools for massive data sets, presenting a synoptic of the main functional and non-functional characteristics of the surveyed tools.

Introduction

I. INTRODUCTION

Data visualization and translation of information are getting vital expertize in today’s trade world [1]. The distinction between what is relevant and what isn't becomes increasingly blurred, as the amount of data available grows. Undoubtedly, the consideration of information visualization tools and dashboards available permits officials to form choices in a brief sum of time.

Data visualization plays a crucial role within the voyage of changing crude information into a more refined form. It helps to present data constructively regardless of what business or vocation you select. One of the essential steps in any business is taking raw data, modeling it, and presenting it in order to draw conclusions [2]. Data visualization tools have the capacity to prepare and show the information in an absorbable manner. It is effective in making solid information reports and dashboards by accumulating enormous amounts of data from distinct sources thereby helping in the discovery of patterns and the formulation of an opinion.

This study distinguishes the data visualization tools on the basis of the format of output, need for coding knowledge and commercial part of the tool [10]. Massive volume of data is displayed in a way that is both accessible and intelligible.

In this paper, we examine several prominent Data Visualization tools, including Tableau, Power BI, Zoho Analytic and others.

II. METHODOLOGY

Exploration of shipment dataset (From Kaggle) is done using six different data visualization tools which are - Tableau, Power BI, Zoho Analytic, Dataiku DSS, Celonis and MS Excel. This dataset contains 10999 observations of 12 variables viz. Warehouse Block, Mode of shipment, Customer rating etc. The dataset is examined to classify the different modes of transportation through which the item was delivered. It is created in order to obtain an exact count of shipments using three different modes of transportation: flight, road, and ship. To learn about client opinions on shipment services, a pie chart is utilized to examine customer ratings by category. Finally, to provide a broad picture of consumer satisfaction, we calculated the average of customer ratings. A line graph is utilized to show the importance of the product in order to count the prioritized items. Furthermore, a bar chart is created to track the number of things dispatched from various warehouses that arrived on time. To aid the users, a histogram is created to provide information on the number of products that fall under a given discount range. Last but not least, a donut chart is created to correlate the number of customer service calls with the number of things that did not arrive at their destination. This is done in order to maintain track or to have a proper check on the customer service.

In this study, we are evaluating and comparing the above-mentioned Data Visualization tools on the basis of various factors such as Free Version availability, Sharing Dashboards, On-boarding/UI feasibility, Customizable Visualization features and charts, Coding, Pricing (Cloud), Pricing (On-Premises), Integration with Other Tools/Scripts, Storage Limit, Data Cleaning, File Upload Size, Customer Support and Data Sources.

III. TOOLS ANALYSIS

A. Tableau

Tableau is an effective tool for Data Visualization that is rapidly growing in popularity. With Tableau, data analysis is very fast, and the dashboards and worksheets that are created are very visually appealing. Tableau's data is straightforward to comprehend by experts at all tiers of a business. A variety of visualizations can be created to display data and showcase insights in a magnificent manner.

1. Pros
   • Complex table calculations can be performed more rapidly with scripting languages such as R and Python.
   • Can handle large amounts of data efficiently.
2. Cons
   • Difficulty in importing custom visualizations.
   • Reports cannot be embedded easily into other applications.

B. Power BI

This analytic and visualization tool by Microsoft lets you analyze and visualize data, extract insights, and share insights within your organization. This tool can handle data from a wide variety of sources. Furthermore, it offers the ability to generate adhoc reports which help in the analysis of data.

1. Pros
   • Reports and dashboards pre-built for SaaS solutions.
   • Data can be scheduled for refresh, without manual intervention after publishing to Power BI web service.
2. Cons
   • Users with the same email domain can only use the dashboard and report sharing feature.
   • The file size of a Power BI report cannot exceed 1 GB.

C.  Zoho Analytic

Zoho Analytic is a business intelligence (BI) platform that was launched in 2009. It provides state-of-the-art analytical solutions that converts uncooked facts into actionable insights. The platform allows users to retrieve data from any data source and visually analyze it to make informed data-driven decisions. Users can also share insights and collaborate.

1. Pros
   • Easy to collaborate and share insights. (Report sharing and collaboration).
   • ZIA (AI Assistant) provides ready-to-use embedded BI solutions.
2. Cons
   • API connections to non-Zoho sources are not reliable.
   • Outdated dashboard and charts design with a timestamp.

D. Dataiku DSS

Dataiku DSS is data science software specifically designed for engineers, analysts and scientists to effectively generate, explore and deploy data products. It is a systematic data science and analysis tool that includes a visual user interface and inbuilt coding. It promotes the use of popular notebooks like Spark, Hive, Python, and R. From wrangling - to analysis to modeling, drag-and-drop can be utilized anywhere within the data-flow prototype process.

1. Pros
   • Ability to build models in Python or R.
   • Very intuitive and easy to use UI.
2. Cons
   • Lags while working with a large amount of data.
   • Lacks proper documentation.

E. Celonis

Celonis Execution Management Platform enable businesses to execute everything from data ingestion to process and task mining to planning and simulation to visual and daily management. It has launched an Execution Management System to eliminate system complexity altogether.

1. Pros
   • Automatization of known issues (Action Flows / Action Engine).
   • Level of granularity of the information.
2. Cons
   • Complexity of PQL programming and inadequate support for it.
   • Using date picker in Celonis is a big hurdle in itself (It wipes out one date on alternating the other)

F. MS Excel

For organizing, filtering, and displaying humongous amount of data, MS Excel is widely utilized. It is most frequently used by accountants or professionals to manage large, heavy datasets, but it could be used by almost everyone. Balance sheets, budgeting, editorial calendars, and data calculators are examples of such applications.

1. Pros
   • It can work with SQL exports.
   • The data is easier to segment.
2. Cons
   • It can be a time-consuming venture.
   • There are no automatic updates. the user's linguistic skills.

IV. MERITS OF DIGITAL DATA VISUALIZATION

1. Information  presented graphically is simple to comprehend.
2. It enables information to be communicated regardless of

3. Large amounts of data can be seen at one go thereby reducing the time taken in reading the information in text or table formats otherwise.

4. This type of data aids in the development of a company decisions.

5. It provides a rapid overview of complicated data that would have taken long time span to comprehend.

6. The user does not rely on others for assistance as the photos are self-explanatory.

7. When it comes to translating, the user may not need any special expertise for interpreting the visualized data.

8. It conveys a more accurate picture of the data than words alone.

9. The user's interaction with the image becomes more user-friendly.

V. RESULTS

Each dashboard is generated and compared to how each software package is perceived by the user in terms of dashboard development. The ultimate outcomes are listed below.

A. Tableau

Tableau lacks features that a full-fledged business intelligence tool would have, such as large-scale reporting, data table creation, and static layouts. The solution's potential to share outcomes is also limited.

But at the same time, it can handle complex calculations with the help of the scripting languages, like R and Python.

B. Power BI

Power BI desktop application is not available for MAC users. They can work only on the cloud environment which requires a strong internet connection otherwise it becomes quite slow to visualize the data.On the contrary, it has ‘summarize’ function through which we can get summary of any visual in Power BI.

C. Zoho Analytic

While visualizing the shipment dataset we discovered that there are insufficient charts and graph templates in the zoho analytics tool. As we wanted to create a heatmap for customer ratings, but owing to a lack of resources(graphs/charts), we were forced to settle for a funnel graph.

On the other hand, it provides a well versed AI Assistant (ZIA), which provides ready-to-use embedded BI solutions and reports.

D. Dataiku DSS

It doesn't have the feature to label the graphs and charts while visualizing the dataset, because the free version restricts it to do so.

However, it has the feature of monitoring and controlling the version. As a result, if something goes wrong, the user can easily roll back to a previous version.

E. Celonis

We utilized PQL (Product Query Language) to create the KPI for the number of things that arrived on time that were dispatched from multiple warehouses. Since there are few resources for PQL on the web, it was challenging to build code for it under particular limits and conditions.

Whereas, with the aid of powerful process mining capabilities, user can analyze the whole process flow with the help of process explorer, case explorer, user analysis tracker, conformance tracker, etc.

F. Microsoft Excel

Along with 0 and 1, an extra row labelled 'Blank' was introduced while visualizing a timely delivered pie chart. This 'Blank' was redundant for the chart. By filtering the blank label from the field button, the 'Blank' had to be manually removed.

Although, it has Pivot Charts, that are used to visually summarize and investigate complex data. It also provides interactive filtering controls directly on the chart, allowing the user to easily study a portion of the data.

Conclusion

In the Big Data field, visualization is a crucial aspect. This paper is focused on various Data Visualization applications, which are utilized in several fields like - Big Data Analytics, commercial, enterprise, websites, mobile platforms, etc., and deliberately examines the tools. As we are able to see, it\'s pretty tough to choose from Power BI, Tableau, Zoho Analytic, Celonis, MS Excel, and Dataiku DSS, but we believe we have done so. However, one has to look at their requirements to determine which one to go along with. For instance, a small and medium commercial enterprise (SMB) employer would possibly be afflicted by licensing price from Tableau regardless of its exceptional features. Power BI and Dataiku DSS are extremely good for hybrid surroundings with computers restrained to customers. Celonis provides larger options for integrations from different tools and scripts. For automated and detailed analysis Zoho Analytic is quite good, considering it has limited sets of chart representations. MS Excel is good for small datasets which are easy to handle manually. Last but not the least, Power BI has great user-friendly UI in the Data Visualization business. The software\'s drag-and-drop functionality, combined with access to over 85 data visualization programs, provides an unrivalled user experience that results in an aesthetically beautiful and useful report.

References

[1] Michael Diamond and Angela Mattia, Jacksonville University, “Data Visualization: an exploratory study into the software tools used by business”, Journal of Instructional Pedagogies, vol. 17 pp. 1-3, November 2015. [2] Mohammed Kemal, University of Liverpool, “Data Visualization Tools in Action Choosing a Visualization Software”, Research-gate., pp. 2, April 2019. [3] Syed M., Noopur G., and Rakesh K. IIIT Bhubaneshwar. “Big Data Vizualization: Tools and Challenges”. pp.709-711, December 2016. [4] Rajeswari C, Dyuti Basu and Namita Maurya, VIT University “Comparative Study of Big data Analytics Tools: R and Tableau”, IOP Conference Series: Materials Science and Engineering, November 2017. [5] Aman Velpula, Anirudh Suhas Venkatayogi and Nitesh Kumar Yadav, “Compare and Contrast Cloud based CRM software: Salesforce.com Vs Zoho crm”, February 2015. [6] Matthew N. O. Sadiku1, Adebowale E. Shadare, Sarhan M. Musa and Cajetan M. Akujuobi, Prairie View A&M University, “Data Visualization”, Volume. 02, December 2016. [7] Soly Mathew Biju and Alex Mathew, “Comparative Analysis Of Selected Big Data Analytics Tools”, Journal of International Technology and Information Management, Volume 26, December 2017. [8] Arockia Panimalar.S, Komal M.Khule, Karthika.S, Nirmala Kumari.T, “Data Visualization Tools and Techniques For Datasets In Big Data”, Sri Krishna Arts and Science College, International Research Journal of Engineering and Technology (IRJET), Volume: 04, August 2017. [9] U. S. Junghare, Shri Shivaji Science college, “Analytical Evaluation towards Big Data Visualization Applications, Tools and It’s Issues”, The International journal of analytical and experimental modal analysis, September 2021. [10] Leena Gokhale and Kirti Nilesh Mahajan,Bharati Vidyapeeth Deemed University, “Comparative Study of Data Visualization Tools” Research gate., pp. 48-49, May 2020.

Copyright

Copyright © 2022 Aryamaan Pandey, Ishita Sharma, Akshat Sachan, Dr. P Madhavan. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.