Knowledge

How we use AI chatbot for Data Analytics

Artificial Intelligence (AI): Yes, we use it (it helps augment data analytics). No, it won’t take over our jobs (but it will continue to make what we do more efficient, for our clients).

May 23, 2023

Open AI picture displaying Chat GTP

Use AI to leverage your success as a data scientist

Artificial Intelligence (AI): Yes, we use it (it helps augment data analytics). No, it won’t take over our jobs (but it will continue to make what we do more efficient, for our clients).

  • Artificial Intelligence (AI) and AI-powered software is not a new phenomenon, especially when it comes to data analytics.

  • AI-powered tools are the backbone of data analytics. If you’ve used analytics software such as Qlik Sense, Microsoft PowerBI, Tableau (Notitia works with all of these and more, in partner with our clients) you’ve already seen AI in action.

AI has assisted Australian industries, prior to the rocket infiltration of AI in 2023. Take healthcare as an example, where machine learning has been used in diagnostics and treatment since the 60s.

Remember when Australian companies first started using chatbots to interact with their customers? According to Stackchat, Australian Broadcasting Corporation launched its NewsBot on Melbourne Cup day in 2016, with AGL Energy Ltd. revealing chatbot Alfie in 2017, and Commonwealth Bank’s virtual assistant Ceba making its appearance in 2018.

What’s new for data analytics: the AI “workplace assistant”

As AI evolves, we expect to see it improve accuracy and “speed up” more manual tasks.

Leveraging AI chatbot to cut down processes and speed up tasks, has been well documented.

Like many other sectors, Notitia has experienced the benefit of AI chatbots (like OpenAI’s ChatGBT, Microsoft’s Bing Chat and Jasper) first-hand across our data services. Our team has spent time understanding the way that they can harness AI technology as a “virtual assistant” to their work.

Here’s how AI is creating day-to-day efficiencies for Notitia’s data analysts, right now.

Saving time + getting more done for our clients:

The number one benefit is saving time. We have multiple examples of this throughout our business, which include testing and creating code, assisting in research and developing training resources for our people and clients.

The teammate who “gets it” 70% of the time

If you work remotely, at different times to your team, or if you work in a siloed role - AI is the workmate that you always wished you had. Have 1 - 1000 questions? Need to bounce ideas off something that 7/10 times will give you a highly intelligent answer? AI takes pressure off your teammates, and cuts out the “web search” middle man.

It helps if you are already an SME. You need to have enough understanding to determine what is “correct” and “incorrect” in the AI answer. This is well documented by users worldwide, and our team has also confirmed this.

Ideas generation & training resources

If you need some inspiration, or an “extra set of eyes' on a topic, AI can give you a framework to start a task, or give you another perspective from a different angle or POV.

Lachlan Hamilton, Notitia’s Analytics Lead, often provides training to the Notitia team and for our clients.

He says that AI is a great starting point for businesses who want to upskill their people in data literacy and produce learning resources.

“I’ve used AI for inspiration and as a research tool, to create training content for different audiences, at their specific level of understanding and technical capability,” he says.

“While a lot of training is hands-on in a classroom or 1:1 setting, using AI is a great way to put that “beginner level” hat on to create introductory content and tailor the framework of educational resources.”

Using AI chatbot to write code

Lachlan says AI is like having an extra teammate who can help to generate ideas or a framework and is highly proficient at any coding language.

“Using AI will help if you already know the basics, such as the approach that might solve your problem, along with some understanding of code to be able to check it,” Lachy says.

“In writing code, best practice is to first write it out in pseudocode, as plain language.

“Using AI, you can just write the pseudocode directly into the software and it will do all the work, to turn it into code.”

Lachy says that AI will “get it wrong the first time”. So be mindful that a human will need to check it and change it. But at the very least, it will give you a much quicker path to the finished result.

“Now, I can just ask Chat GPT ‘hey, write me some SQL to create a table for this data’ and I might just paste a little snippet of data and it'll just go and do it,” Lachy says.

“What might have taken half an hour or 45 minutes, you can now do in five minutes.”

“It’s probably going to get it wrong the first time, but you can correct it and say ‘change the type of that column from a whole number to a decimal number’ for example, and it'll just do it.

Remember to use AI with caution

The exponential growth of digital data and technology, the regulations to protect users and consumers often lag behind (insert link to cyber security article). AI is no different, in fact, Elon Musk and researchers worldwide have called for a “pause” on the technology.

Two-thirds of Australians say there are not enough laws or regulations preventing artificial intelligence from being abused. A report by consultancy firm KPMG and the Australian Information Industry Association (AIIA), the industry peak body, found Australians expect AI to be regulated and want an independent regulator to monitor the technology as it booms into mainstream society.

It is also important to note that anything you share with AI tools such as ChatGPT is retained to train the model, with Samsung learning this lesson the hard way after leaking top secret data.

(Image: Dado Ruvic/Reuters)

Notitia services & expertise

Want to know more about how Notitia can help with your data strategy? Book in a 1:1 chat with our team.

We use our combined expertise to provide a solution with tangible outcomes for your business and people.

Analytics:
  • Problem Solving
  • Training
  • Dashboards
  • BI Analytics
  • Solution Architecture & Implementation
  • Data Lake
Data Strategy:
  • Strategy Workshop Sessions
  • Technical Roadmaps to achieve business goals
  • Data Literacy skills
  • Data Quality
  • Data Governance
  • Cyber Security
  • Data Dictionary
  • Training & process implementation in your business
Web & Software Development:
  • Simple, efficient  & understandable solutions across front  + back end application development
  • Modern Cloud Technologies
Design (meet our team here):
  • Beautiful dashboard design (we work through this with you in an iterative process)
  • UI/UX Design + Prototyping
  • Graphic Design
  • Website and Application Design + Development
  • Brand Identity

Notitia's Data Quality Cake recipe