Feedback from last week led to the development of subquestions:
How does the absence of alt text in ePub files impact visually impaired users access to information, and can machine learning effectively generate appropriate alt text?
And/or:
What is the effect of missing or incorrect headings in ePub files on accessibility for assistive technology users, and can machine learning help automate the correction of heading structures without destroying any crucial information?
Data collection from both an “unofficial” source and a legitimate subscription-based platform, with a substantial dataset available.
Objectives for Next Week:
Completion of the Exploratory Data Analysis (EDA) by the upcoming Friday is the primary goal.
Remarks:
Considering the complexity to each subject, is a focus on of the question better over an attempt to address both?
Is there a preference for avoiding data from unofficial sources, or is the emphasis on the data’s quality and relevance?
@maartenmarx
Recap:
How does the absence of alt text in ePub files impact visually impaired users access to information, and can machine learning effectively generate appropriate alt text?
And/or:
What is the effect of missing or incorrect headings in ePub files on accessibility for assistive technology users, and can machine learning help automate the correction of heading structures without destroying any crucial information?
Objectives for Next Week:
Completion of the Exploratory Data Analysis (EDA) by the upcoming Friday is the primary goal.
Remarks: