YZ Sözlüğü
Yapay Zekanın tam sözlüğü
Sentiment analysis by lexicon
Approach using dictionaries of words pre-labeled with their polarities to calculate the overall score of a text.
Sentiment Analysis with Machine Learning
Supervised methods using classical algorithms such as SVM, Naïve Bayes, or Random Forest to classify sentiments.
Sentiment Analysis with Deep Learning
Use of deep neural networks (LSTM, GRU, Transformers) to capture complex contextual relationships in text.
Aspect-based sentiment analysis
Identification of specific aspects of a product/service and analysis of the sentiment associated with each aspect individually.
Fine-grained emotion detection
Classification beyond positive/negative to identify specific emotions such as joy, anger, fear, surprise, disgust, or sadness.
Analysis of sarcasm and irony
Detection of expressions where the literal meaning differs from the intended meaning, requiring advanced contextual understanding.
Multilingual sentiment analysis
Adapted techniques for processing and analyzing sentiment in multiple languages with cross-lingual or specific models.
Real-time sentiment analysis
Systems optimized for instant processing of continuous data streams such as social media or live comments.
Multimodal sentiment analysis
Integration of multiple modalities (text, images, audio, video) for a more comprehensive and nuanced sentiment analysis.
Domain-oriented sentiment analysis
Adapting sentiment analysis models to the specific vocabularies and expressions of particular domains (medical, financial, etc.).
Comparative sentiment analysis
Identification of preferences and comparisons between different entities or options within the same text.
Implicit sentiment analysis
Detection of sentiments expressed indirectly without explicit polarity words, requiring contextual inference.