CBSE Class 10 Artificial Intelligence Curriculum Vocabulary Flashcards

A complete set of vocabulary flashcards covering basic and advanced AI concepts, modeling, evaluation metrics, computer vision, and Natural Language Processing based on the Class 10 Facilitator Handbook.

AI Project Cycle

The cyclical process followed to complete an AI project, consisting of stages: Problem Scoping, Data Acquisition, Data Exploration, Modeling, Evaluation, and Deployment.

Problem Scoping

The initial stage of the AI project cycle where the goal is set by defining the problem that needs to be solved.

Data Acquisition

The process of collecting data from various reliable and authentic sources to serve as the base for an AI project.

Data Exploration

A stage in the AI project cycle where visual representations like graphs and maps are used to interpret patterns in acquired data.

Modeling

The stage where a suitable algorithm is researched, selected, and tested to achieve the project goal.

Deployment

The final stage of the AI project cycle ensuring the successful integration and operation of AI solutions in real-world environments.

Statistical Data Domain

A domain of AI related to data systems and processes where the system collects data sets to derive meaning for decision-making.

Computer Vision (CV)

A domain of AI that enables machines to analyze visual information (photographs, videos) and predict decisions about it.

Natural Language Processing (NLP)

A branch of AI that enables computers to analyze, understand, and process human languages (spoken or written).

Ethical Frameworks

Step-by-step guidance used to ensure that problem-solving choices do not cause unintended harm and align with certain moral principles.

Bioethics

A sector-based ethical framework used in healthcare and life sciences to address issues like patient privacy and data security.

Non-maleficence

An ethical principle of avoiding causing harm or negative consequences, prioritizing the path of least harm.

Beneficence

The ethical principle of promoting and maximizing the well-being and welfare of individuals and society.

Artificial Intelligence (AI)

The umbrella term for any technique that enables computers to mimic human intelligence using algorithms and data.

Machine Learning (ML)

A subset of AI that enables machines to improve at tasks with experience by learning from new data without explicit programming.

Deep Learning (DL)

A specialized learning approach and subset of ML that uses vast amounts of data and multiple algorithms (neural networks) to train itself.

Features

The columns of a dataset that describe characteristics used by an AI model to make predictions.

Labels

Specific information attached as a tag to data used as the target variable for predictions in a model.

Supervised Learning

A machine learning approach where a model is trained using labeled data, acting like a student learning from a teacher.

Unsupervised Learning

A machine learning approach where a model independently discovers patterns and relationships within an unlabeled dataset.

Reinforcement Learning

A learning approach where a machine learns through trial-and-error using a reward mechanism to maximize positive outcomes.

Classification Model

A supervised learning model used to categorize data into discrete, predefined classes or labels.

Regression Model

A supervised learning model that works on continuous data to predict numerical values like price, age, or temperature.

Clustering

An unsupervised learning method that finds similarities between objects and groups them into clusters without predefined labels.

Association Rule

An unsupervised learning method used to discover interesting relationships between variables in large databases based on patterns.

Artificial Neural Networks (ANN)

Deep learning models modeled on the human brain that automatically extract data features efficiently using interconnected layers of nodes.

Convolutional Neural Network (CNN)

A deep learning algorithm designed to assign importance to objects in an image and differentiate one from another.

Perceptron

A mathematical model of a neural node used for making binary decisions by summing weighted inputs and comparing them to a threshold.

Train-test split

A technique for evaluating model performance by dividing a dataset into a training subset to teach the model and a testing subset to verify it.

Accuracy

An evaluation metric measuring the total number of correct predictions a model makes as a ratio of all predictions.

Error

The difference between a model's prediction and the actual outcome, quantifying how often the model makes mistakes.

Confusion Matrix

A tabular presentation used to summarize the accuracy of a classification model across various actual and predicted classes.

True Positive (TP)

The outcome where the model correctly predicts the positive class, such as identifying a diseased person as having the disease.

False Positive (FP)

The outcome where a model incorrectly predicts the negative class as positive, such as identifying a healthy person as having a disease.

Precision

The ratio of correctly classified positive examples to the total number of examples predicted as positive: $\frac{TP}{TP + FP}$.

Recall

The measure of a model's ability to correctly identify actual positives, also known as Sensitivity: $\frac{TP}{TP + FN}$.

F1 Score

A combined metric used for unbalanced datasets that balances precision and recall: $2 \times \frac{\text{Precision} \times \text{Recall}}{\text{Precision} + \text{Recall}}$.

No-Code AI

Visual AI development platforms that use drag-and-drop features instead of manual programming to build applications.

Low-Code AI

AI development platforms that provide visual interfaces but still require a limited amount of manual coding for customization.

Orange Data Mining

An open-source, no-code data visualization and machine learning toolkit used for building AI workflows.

Mean

A descriptive statistic representing the central value or common average of a set of numbers.

Median

The middle value in a dataset when numbers are ordered from low to high.

Normal Distribution

A symmetrical distribution of values clustered around a central peak, appearing bell-shaped.

Variance

A measurement of how far each value in a data set is spread out from the mean.

Standard Deviation

A calculation representing how widely distributed values are within a dataset relative to the mean.

Outlier

A specific data point that lies at an abnormal distance from other values in a dataset.

Image Processing

A subset of computer vision focused on processing raw input images to enhance them or prepare them for further analysis.

Pixel

The smallest unit of information that makes up a digital picture, standing for "picture element."

Resolution

The number of pixels in an image, often calculated as width multiplied by height (e.g., $1280 \times 1024$).

Pixel Value

A numerical description of a pixel's brightness or color, typically ranging from $0$ (black) to $255$ (white) in an $8$-bit byte image.

Grayscale Image

An image consisting of shades of gray where each pixel has a single byte value representing intensity from $0$ to $255$.

RGB Image

A color image stored using three channels—Red, Green, and Blue—where each pixel is formed by combining intensities from all three planes.

Kernel

A sliding matrix used in convolution to multiply image arrays and enhance specific visual features.

Convolution Layer

The first layer of a CNN responsible for extracting high-level features like edges and colors from an input image.

Pooling Layer

A layer in a CNN that reduces the spatial size of convolved features to make management and processing more efficient.

Lexical Analysis

The initial stage of NLP involving identifying the structure of words and dividing text into sentences and tokens.

Syntactic Analysis

The stage of NLP that checks the grammatical structure of sentences to ensure relationships between words are valid.

Semantic Analysis

The stage of NLP where input text is checked for actual meaning and logical meaningfulness within a sentence.

Pragmatic Analysis

The final stage of NLP that checks sentences for relevance and intended meaning in the real world.

Text Normalization

The process of simplifying complex language into tokens and basic forms to clean up textual data for computer processing.

Tokenization

Dividing sentences into individual units called tokens, which include words, numbers, and special characters.

Stop Words

Frequently occurring words like "and," "the," and "is" that add little value to the context and are often removed during preprocessing.

Stemming

A text processing technique that removes affixes from words to reduce them to a base form, which might not be a meaningful word (e.g., "studi").

Lemmatization

A text processing technique that removes affixes to ensure the resulting root word (lemma) is meaningful (e.g., "study").

Bag of Words

An NLP algorithm that extracts features from text by creating a dictionary of unique words and counting their frequencies.

TFIDF

Short for Term Frequency-Inverse Document Frequency, a metric used to identify the relative value and importance of each word in a corpus.