WriteWise: Elevating English Writing Concept Design
I created a website concept design utilizing a large language model to assist non-native speakers in improving their writing skills.
Challenge
Large language models such as ChatGPT have the potential to complete multiple tasks, but their broad utility limits their ability to be applied to specific tasks effectively.
Outcome
Approach
How might we assist English learners in utilizing ChatGPT to enhance their writing skills?
The problem space can be divided into three smaller issues, which I attempt to address through academic research, user interviews, and competitive analysis.
Academic Research
English writing improvement is a widely researched field that encompasses various disciplines:
Insight 1: As learners improve, their needs evolve
English writing improvement is a widely researched field that encompasses various disciplines:
Insight 2: Within distinct stages of every writing endeavor, diverse needs emerge
The process of composing a piece of writing can be divided into different stages and need different instructions.
These stages may repeat or backtrack to achieve a satisfactory outcome.
Competitive Analysis
Writing improvement is a fiercely competitive market. Despite the absence of AI integration in existing products within this field, they exhibit the following drawbacks:
Instead of enhancing writing skills, these tools focus primarily on proofreading essays.
Without AI assistance, these tools are limited to correcting grammar and spelling errors, lacking the capability to improve the content and structure of essays.
User Interview
What are learners' own perspectives?
After reviewing academic literature and competitors in the field, I sought to gather first-hand insights from my target audience about their challenges and needs.
I conducted a pre-interview survey on English writing experience and recruited 6 college and graduate students from a pool of 21 respondents to simulate a rough stratified sampling.
These 6 participants, with an average age of 22, all have experience in English writing, with IELTS scores ranging from 5 to 8 (converted).
Key Takeaways
Blue is for neutral emotions, red for negative, and green for positive
Extracting Target Users from the Interview:
Design Decisions
Solution 1: Enhance Writing Quality with Comprehensive Guidance on Expression and Organization
Offer suggestions encompassing grammar, spelling, as well as expression elements like vocabulary, phrasing, or sentence clarity, highlighted in distinct colors for easy identification.
Users can also utilize AI to request essay rewriting, addressing issues like lengthiness or lack of clarity in the main points.
Solution 2: Generate Automated Notes for Efficient Review
Tailored suggestions are available throughout the entire writing process
Users can customize the writing context to receive writing prompts, offering valuable insights without enabling direct plagiarism
Other projects
Ceribell: Medical Notification System Research & Design
I contributed to designing the medical notification system to improve communication in healthcare institutions, resulting in a user satisfaction score of 8.6 in the pilot version.
ClipEase: Smart Clipper User Research
I assisted a Chinese AI-driven haircutting solution provider in exploring the U.S. market through in-depth interviews, sidewalk surveys, and quantitative analysis.
Beichoo: Recruitment Website Function Design
I designed two functions to streamline recruiters' tasks at Beichoo, which increased the website's monthly active users tenfold, from 1,857 to 18,314.