- Client:
- Professional and personal growth.
- Project(s):
- Gaining skills that I apply in projects as needed.
- Role:
- Acquiring and applying skills strategically to enhance project outcomes.
About the project(s)
Using AI in my daily UX practice with data visualisation and sentiment analysis.
In my daily UX practice, I utilise tools like AWS Amazon Comprehend whenever feasible and aligned with business requirements to analyse text in real-time. There are a lot of tools in the market and companies are starting to create similiar ones for internal use too. These tools allow you to extract key phrases, detecting dominant languages, analysing sentiment, and identifying syntax.
Amazon Comprehend, powered by AWS, is a Natural Language Processing (NLP) tool that uses machine learning to uncover insights and relationships within text of UTF-8 encoded text documents.
For example, starting with an online review, such as one from BT, we can quickly extract key sentiments and insights using Amazon Comprehend, as shown in the several example images.
Key insights can be uncovered by:
- Extracting key phrases.
- Analysing entity insights.
- Identifying key sentiments.
- Gaining insights from targeted sentiment.
Picking up an online review.
Analysing an online review by picking up type of analysis.
Picking up on entities insights. specific people, places, brands, products, or concepts mentioned in a text. They are often extracted using Named Entity Recognition (NER) and analyzed to determine sentiment related to them.
Picking up on key phrases.
Looking at targeted sentiment.
Getting the final sentiment analysis report.
How does it work?
- The DetectSentiment operation returns an object that contains the detected sentiment and a SentimentScore object.
- The BatchDetectSentiment operation returns a list of sentiments and SentimentScore objects, one for each document in the batch.
- The StartSentimentDetectionJob operation starts an asynchronous job that produces a file containing a list of sentiment, sand SentimentScore objects, one for each document in the job.
- The sentiment determination returns one of the following categories ina report:
- 😊Positive – The text expresses an overall positive sentiment.
- 😠Negative – The text expresses an overall negative sentiment.
- 😔Mixed – The text expresses both positive and negative sentiments.
- 😑Neutral – The text does not express either positive or negative sentiments.
- I gather these reports to create empathy maps and communicate insights to the stakeholders.
How do I use these insights into user experience design and product development?
This process allows me to use insights quickly, which otherwise would take longer to extract and, apply them to empathy maps and user journey maps, allowing me to understand user challenge and redefine the problem statement further with background, objective data.
I can create UX deliverables such as empathy maps, user journeys and graphs to explain the user experience and tell a story (storytelling) easily to stakeholders with reliable background data. Amazon Comprehend allows you to spot factors, key trends and sentiment very quickly but it is still required a human overview to discern what is important and irrelevant for a particular project, the storytelling and filter necessary to make sense of information a useful piece of work with value for an organisation.
What am I currently doing?
Creating a chatbot with Mistral! You can see what I have done so far here.
Learning and practising with AI, ML, Data and Data Visualisation.
Stay tuned for updates and enhancements to this portfolio page. I am always exploring, experimenting, and refining my skills through coding, GitHub, and writing about topics such as:
- AI product design & UX research for AI models.
- Conversational UX & voice AI.
- Explainable AI (XAI) & ethical UX.
- Data visualization & AI interaction design.
- Human-AI collaboration & automation.