Diagnosing a Text Classification System Failure
A system is designed to classify customer feedback. It first uses a component to convert text into a rich numerical representation. A second component then uses this representation to assign a 'positive' or 'negative' label. Given the scenario described below, identify which component is the most probable source of the error and justify your reasoning.
0
1
Tags
Ch.1 Pre-training - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Analysis in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
Mathematical Notation for a Classifier in a Text Classification System
Example of Text for Polarity Classification
A system is designed to classify customer reviews as 'positive' or 'negative'. The system operates in two stages: first, a component converts the review's text into a detailed numerical vector that captures its meaning. Second, another component takes this vector as input, calculates a score for the 'positive' label and a score for the 'negative' label, and then outputs the label with the higher score. If this system processes the review 'The service was impeccable!', which component is directly responsible for the final decision to label the review as 'positive'?
A system is designed to determine if a movie review is 'positive' or 'negative'. Arrange the following steps in the correct logical order, from the initial input to the final output.
Diagnosing a Text Classification System Failure