Answer: Natural language processing (NLP)

## Detailed Explanation This scenario describes a customer service team developing an application to analyze text-based customer feedback and automatically classify it into predefined categories (product quality, customer service, and delivery experience). This is a classic application of **Natural Language Processing (NLP)**, which is a core AI concept focused on enabling computers to understand, interpret, and manipulate human language. ### Why Option B (Natural Language Processing) is Correct: 1. **Text Analysis**: The scenario involves processing unstructured text data from customer feedback, which is the primary domain of NLP. 2. **Text Classification**: Automatically categorizing text into predefined groups (product quality, customer service, delivery experience) is a fundamental NLP task known as text classification or sentiment categorization. 3. **Language Understanding**: The application needs to comprehend the meaning and context of customer comments to assign appropriate categories, requiring NLP techniques like tokenization, semantic analysis, and machine learning models. ### Why Other Options Are Not Suitable: - **Option A (Computer Vision)**: Computer vision deals with analyzing visual data like images and videos. This scenario involves text analysis, not visual content. - **Option C (Recommendation Systems)**: Recommendation systems suggest items based on user preferences (e.g., "customers who bought X also bought Y"). This scenario involves categorization, not personalized recommendations. - **Option D (Fraud Detection)**: Fraud detection typically involves identifying anomalous patterns in transactional data to prevent fraudulent activities. While it can use AI, this scenario is specifically about text classification for customer feedback analysis. ### Key NLP Concepts Involved: - **Text Preprocessing**: Cleaning and preparing text data for analysis - **Feature Extraction**: Converting text into numerical representations (e.g., word embeddings) - **Classification Algorithms**: Using supervised learning models to assign categories based on training data - **Sentiment Analysis**: Potentially identifying positive/negative sentiment within each category This application could leverage AWS AI services like Amazon Comprehend, which provides pre-trained NLP capabilities for text classification, entity recognition, and sentiment analysis, making it easier to implement such solutions without building models from scratch.

Author: LeetQuiz Editorial Team