04a text classifiers sentiment evaluation SLP

Text Classification

The process of categorizing text into predefined classes.

Naive Bayes Classifier

A classification algorithm based on Bayes' theorem, assuming independence between predictors.

Sentiment Analysis

The detection of attitudes expressed in text, typically as positive or negative.

Precision

The percentage of true positive predictions relative to the total positive predictions made by the model.

Recall

The percentage of true positive predictions relative to the total actual positives in the data.

F-measure

A combined measure of precision and recall, often used as a single metric for evaluation.

Bag of Words

A representation of text data where all words are treated as individual tokens, ignoring grammar and word order.

Maximum Likelihood Estimate

A statistical method for estimating parameters of a probabilistic model by maximizing the likelihood function.

Laplace Smoothing

A technique used to avoid zero probabilities in maximum likelihood estimation by adding a small constant, usually 1.

Confusion Matrix

A table used to evaluate the performance of a classification model, showing true positives, false positives, true negatives, and false negatives.

Macroaveraging

A method of averaging performance metrics by computing them for each class separately and then averaging the results.

Microaveraging

A method of averaging performance metrics by computing global counts across all classes before calculating precision and recall.

Toxicity Classification

The task of detecting harmful language such as hate speech, abuse, or harassment.

Model Card

A documentation tool for machine learning models that includes details about the model's training, performance, and intended usage.

Multinomial Naive Bayes

An extension of the Naive Bayes classifier that is suitable for discrete data, such as word counts for text classification.

Supervised Machine Learning

A type of machine learning where models are trained on labeled data to predict outcomes for new, unseen data.

Negation Handling

Methods used in sentiment classification to accurately interpret the effect of negations on sentiment words.

Unknown Words

Words that appear in the test set but not in the training set, which are typically ignored in sentiment analysis.

Stop Words

Very common words in a language (such as 'the' or 'and') that may be excluded from text analysis in some classification tasks.