I wanted to test out which are the best AI data analysis tools to help me with user research data. I decided to check which are most popular tools by visiting the HDRobots data analysis page. I extracted the top 10 most popular ones and tested each one of them.
What did I learn? Only the first and the third most popular AI data analysis tools impressed me.
Quick Experiment Breakdown
- Used a uniform dataset: 2023 tennis matches from http://data.world.
- Tasks for AI: Identify top 10 players by victories and calculate top 10 by winning percentage. The latter was more complex.
- Exclusions: PowerBI & MonkeyLearn (needed corporate accounts); RoseAI, InfraNodus, DataRobot (format/feature limitations).
- Effectively, only 5 AI tools were applicable for the tennis dataset analysis.
#10 Rose AI
@rosedotai assists with finding, cleaning, visualizing, and transforming data. The user interface was confusing and lacked sufficient help. The http://Rose.AI data chatbot is not fully operational, and we were unable to use it. Additionally, it does not support uploading XLSX or CSV files with time-series data arranged horizontally. Consequently, I were unable to test Rose AI with our tennis dataset.
#9 InfraNodus
InfraNodus is specialized in analyzing text data. It converts text into graphs where you can identify relations, find relevant words and generate insights using AI. It has a 14-day free trial and then it starts from 9 EUR per month.
#8 Polymer
@polymersearch is a business intelligence tool that uses AI to visualize data. It has a chatbot but it doesn’t support creating graphs and it couldn’t answer the 2nd question. Polymer requires manual work to analyze the data. It has a 14-day free trial, then $20 per month.
#7 Akkio
@AkkioHQ is an easy to use data analysis tool. It has example data to try out the tool. It also has a data prediction feature. The tool successfully answered the first question, however it couldn’t calculate the answer for the second one. It features a 14-day unlimited free trial, then $49 per user per month.
#6 DataRobot
DataRobot builds a customized AI model based on your data with the goal of predicting future trends. It’s geared towards data scientists, AI Researchers and ideal for collaborative use. I couldn’t managed to set it up to predict next years number 1 tennis player.
#5 Kanaries
Kanaries automates data exploration and generates charts and dashboards. It requires adding a bank card to access the free trial and it costs $10 per month after that if you don’t cancel. During our testing, the tool created unrelated charts and was unable to analyze the data effectively. The interface was a bit hard to use.
#4 MonkeyLearn
MonkeyLearn focuses on customer feedback analysis, such as net promoter Scores (NPS) and satisfaction surveys. With no free tier available, testing requires an account. Its popularity hints at its effectiveness, although I couldn’t personally test it.
#3 Julius
Julius is an easy-to-use data analysis tool. It successfully answered the two tennis questions and created accurate graphs. I was impressed with its accuracy and intelligence. I would definitely recommend Julius for AI data analysis. It offers 15 free queries per month. After that, the price starts at $17.99 for 250 messages per month.
#2 PowerBI
PowerBI by Microsoft is an enterprise business intelligence tool with a lot of functionalities. However, it requires a work account, and I couldn’t test it out because I didn’t have one. It’s the second most popular tool according to HDRobots’ ranking algorithm.
https://microsoft.com/en-us/power-platform/products/power-bi/
#1 Data Analysis GPT
Developed by OpenAI, this was the easiest-to-use data analysis tool. It quickly and successfully completed the two tasks, requiring no setup at all. No wonder it’s the most popular one. Definitely recommended! The only caveat is that it requires a ChatGPT subscription, which costs $20.
https://chat.openai.com/g/g-HMNcP6w7d-data-analyst…
Summary
Out of the five tools that I was able to test for data analysis on the tennis dataset, only Data Analysis GPT and Julius successfully completed the two tasks. Akkio and Polymer couldn’t answer the second question, which required calculating the winning percentage.
Recommendation
Use Data Analysis GPT if you have a ChatGPT subscription, or Julius if you don’t. These are currently the best two data analysis tools.
That’s it. See the original Twitter Thread.
