Spam Filtering/ Text classification

language identification

the process of determining the language of a given text

supervised learning

a type of machine learning where a model is trained using labeled data

training/ testing data

data used to train a machine learning model and evaluate its performance

document classification

the task of assigning a document to one or more predefined categories

binary classification

a classification task with two possible outcomes

multi-class classfication

a classification task with more than two possible categories

Bayes Rule

a mathematival formla used to update probabilities based on new evidence

Naive Bayes

a probabilistic classifier based on Bayes’ Theorem with an assumption of independence among features

logistic regression

a statistical model used for binary classification problems

false positives

incorrectly identifying a non-relebant instance as relevant

false negatives

failing to identify a relevant instance

character n-gram

a sequence of N consecutive characters used in text analysis

spam

unwanted or unsolicited messages, typically emails

spam-filter

a system used to detect and black spam messages

blacklist

a list of entities that are blocked from accessingg a system or service

whitelist

a list of approved entities that are allowed access to a system or service

rule-based filtering

a spam detection approach usingh manually crafted rules

spam probability

the likelihood that a given message is spam

statistical filtering

a spam detection method based on statstical analysis of message content

hand crafted features

features manually designed by experts for machine learning models

kitchen sink features

an approach that includes many featires without filtering for relevance

sparse features

features that have many zero or missing values

dense features

features that have mostly nonzero values and provide rich information