Artificial Intelligence | Internboot Assesments

An AI internship is a short-term, structured opportunity for students or early-career professionals to gain hands-on experience in Artificial Intelligence domains such as machine learning (ML), deep learning (DL), data science, natural language processing (NLP), computer vision, and more.

What You Learn in an AI Internship

Programming: Python (most common), with libraries like NumPy, Pandas, Scikit-learn, TensorFlow, PyTorch
Mathematics: Linear algebra, probability, statistics
ML/DL Techniques: Supervised, unsupervised learning, CNNs, RNNs, transformers
Projects: Model training, data preprocessing, real-time predictions, AI application development
Tools: Jupyter, Git, Docker, APIs, cloud platforms (AWS, GCP, Azure)

Important Notice:
Once you start the quiz, you will not be able to pause, exit, or restart it. Please ensure you are ready before beginning.

Artificial Intelligence L1

1 / 100

Q1. What is the primary goal of Artificial Intelligence?

a) Increase storage capacity

b) Develop software tools

c) Enable machines to mimic human intelligence

d) Create faster processors

2 / 100

Q2. Who is known as the father of Artificial Intelligence?

a) Alan Turing

b) Charles Babbage

c) John McCarthy

d) Marvin Minsky

3 / 100

Q3. Which of the following is a type of AI that reacts to input but does not store memories?

a) Limited Memory AI

b) Reactive Machine

c) General AI

d) Self-aware AI

4 / 100

Q4. In which year was the term "Artificial Intelligence" first coined?

a) 1943

b) 1956

c) 1965

d) 1980

5 / 100

Q5. Which of these is NOT a branch of Artificial Intelligence?

a) Robotics

b) Machine Learning

c) Compiler Design

d) Natural Language Processing

6 / 100

Q6. What does “Turing Test” evaluate?

a) Speed of computers

b) Accuracy of algorithms

c) Intelligence of a machine

d) Data transfer rate

7 / 100

Q7. What type of AI system can understand, learn, and apply intelligence to solve any problem?

a) Narrow AI

b) Strong AI

c) Weak AI

d) Applied AI

8 / 100

Q8. What is an agent in AI?

a) A compiler

b) A robot

c) A program that perceives and acts

d) A neural network

9 / 100

Q9. Which of the following is an example of Artificial Intelligence in use?

a) Electric fan

b) Smart Assistant (e.g., Alexa)

c) Digital watch

d) Calculator

10 / 100

Q10. Which environment does not change while the agent is deliberating?

a) Episodic

b) Static

c) Dynamic

d) Discrete

11 / 100

Q11. What is Machine Learning?

a) Programming computers manually

b) A type of antivirus software

c) Making machines learn from data

d) Copying human behavior exactly

12 / 100

Q12. Which of the following is a type of machine learning?

a) Structural learning

b) Rule-based learning

c) Supervised learning

d) Logical programming

13 / 100

Q13. Which algorithm is used for classification tasks?

a) Linear Regression

b) K-Nearest Neighbors

c) K-Means

d) PCA

14 / 100

Q14. In supervised learning, what does the algorithm learn from?

a) Unlabeled data

b) Feedback

c) Labeled data

d) Clustering patterns

15 / 100

Q15. What is the main goal of regression in ML?

a) Classification

b) Clustering

c) Predicting continuous values

d) Creating labels

16 / 100

Q16. Which of the following is an unsupervised learning algorithm?

a) Decision Trees

b) Linear Regression

c) K-Means

d) Random Forest

17 / 100

Q17. Overfitting occurs when the model:

a) Performs well on new data

b) Is too simple

c) Learns noise in the training data

d) Uses regularization

18 / 100

Q18. What is a feature in machine learning?

a) A test sample

b) A training model

c) An input variable

d) An algorithm

19 / 100

Q19. Which Python library is commonly used for ML tasks?

a) Flask

b) NumPy

c) TensorFlow

d) Django

20 / 100

Q20. What does a confusion matrix evaluate?

a) Data transformation

b) Algorithm accuracy

c) Classification performance

d) Feature extraction

21 / 100

Q21. What is a neural network inspired by?

a) Internet routers

b) Human brain

c) Animal instincts

d) Genetic codes

22 / 100

Q22. What is a neuron in a neural network?

a) A memory block

b) An input layer

c) A computation unit

d) A data structure

23 / 100

Q23. What does the hidden layer in a neural network do?

a) Stores memory

b) Makes decisions

c) Performs computation and transformation

d) Sends data to the output

24 / 100

Q24. Which activation function is most commonly used in hidden layers?

a) Sigmoid

b) ReLU

c) Softmax

d) Tanh

25 / 100

Q25. What is backpropagation used for in neural networks?

a) Forward feeding

b) Weight initialization

c) Error calculation and weight updating

d) Regularization

26 / 100

Q26. What is the role of weights in a neural network?

a) Control learning rate

b) Combine features

c) Represent the strength of connections

d) Store model accuracy

27 / 100

Q27. What is the purpose of the loss function?

a) To classify the data

b) To evaluate how well the model performs

c) To store predictions

d) To increase model complexity

28 / 100

Q28. In a feedforward neural network, information flows:

a) Forward and backward simultaneously

b) Only backward

c) Only forward

d) In a circular loop

29 / 100

Q29. What is an epoch in deep learning?

a) A layer in a network

b) One forward pass

c) One full cycle of training on entire dataset

d) A feature scaling technique

30 / 100

Q30. What does the Softmax function output?

a) A binary result

b) Raw logits

c) A probability distribution

d) A loss value

31 / 100

Q31. What does NLP stand for in the context of AI?

a) Neural Linguistic Programming

b) Natural Language Processing

c) National Language Processing

d) Natural Logic Program

32 / 100

Q32. Which of the following is a common preprocessing step in NLP?

a) Image segmentation

b) Tokenization

c) Backpropagation

d) Gradient clipping

33 / 100

Q33. What is the purpose of stemming in NLP?

a) Remove punctuation

b) Convert words to their root form

c) Translate text

d) Classify sentiment

34 / 100

Q34. Which algorithm is used for word embedding?

a) K-Means

b) Word2Vec

c) Apriori

d) Decision Tree

35 / 100

Q35. What is TF-IDF used for?

a) Image compression

b) Feature scaling

c) Word importance weighting

d) Clustering

36 / 100

Q36. In NLP, what does POS tagging refer to?

a) Preposition Sorting

b) Part-of-Speech Tagging

c) Position Optimization Syntax

d) Phrase Ordering System

37 / 100

Q37. Which NLP task involves detecting if a sentence expresses a positive or negative opinion?

a) Machine translation

b) Sentiment analysis

c) Text summarization

d) Speech synthesis

38 / 100

Q38. Which model is typically used in machine translation?

a) CNN

b) GAN

c) Seq2Seq

d) Naive Bayes

39 / 100

Q39. What is lemmatization?

a) Removing stop words

b) Breaking a sentence into words

c) Converting a word to its base dictionary form

d) Merging duplicate words

40 / 100

Q40. Which of the following is a transformer-based model?

a) Naive Bayes

b) BERT

c) LSTM

d) KNN

41 / 100

Q41. What is the primary goal of computer vision?

a) Understand computer hardware

b) Translate code

c) Enable machines to interpret visual data

d) Perform network analysis

42 / 100

Q42. Which technique is commonly used for object detection in images?

a) PCA

b) YOLO

c) K-means

d) DBSCAN

43 / 100

Q43. What is the output of edge detection in images?

a) Blurred image

b) Image with enhanced color

c) Image highlighting object boundaries

d) Grayscale image

44 / 100

Q44. What does CNN stand for in the context of computer vision?

a) Central Neural Network

b) Convolutional Neural Network

c) Computational Node Network

d) Core Network Node

45 / 100

Q45. Which filter is commonly used for edge detection?

a) Gaussian

b) Sobel

c) Median

d) Box

46 / 100

Q46. Which of the following is an open-source computer vision library?

a) NumPy

b) Pandas

c) OpenCV

d) Flask

47 / 100

Q47. Image classification involves:

a) Generating random images

b) Assigning a label to an image

c) Enhancing image colors

d) Making image copies

48 / 100

Q48. What is the primary advantage of using CNNs over traditional neural networks in vision?

a) Better at text processing

b) Lower memory usage

c) Capture spatial hierarchies

d) Require no training

49 / 100

Q49. Which of the following is used for image segmentation?

a) RNN

b) SVM

c) U-Net

d) LSTM

50 / 100

Q50. Which task involves identifying and locating multiple objects in an image?

a) Classification

b) Detection

c) Captioning

d) Translation

51 / 100

Q51. What does precision measure in a classification model?

a) The ability to identify only the relevant data points

b) The ability to identify all relevant data points

c) The ratio of false positives to total predictions

d) The number of correct predictions over total predictions

52 / 100

Q52. What is a confusion matrix used for?

a) Data cleaning

b) Feature selection

c) Evaluating classification models

d) Normalizing datasets

53 / 100

Q53. Which metric is most suitable for evaluating a model on imbalanced data?

a) Accuracy

b) Recall

c) ROC-AUC

d) Mean Absolute Error

54 / 100

Q54. What does F1-score represent?

a) Precision + Recall

b) Accuracy

c) Harmonic mean of precision and recall

d) Geometric mean of recall and accuracy

55 / 100

Q55. What is cross-validation mainly used for?

a) Feature scaling

b) Testing on the same training set

c) Model evaluation on different data subsets

d) Cleaning missing data

56 / 100

Q56. What is the purpose of regularization in model training?

a) Increase model complexity

b) Remove noise from the dataset

c) Reduce overfitting

d) Improve learning rate

57 / 100

Q57. Which of the following is a regularization technique?

a) Max pooling

b) Dropout

c) Cross-validation

d) ReLU

58 / 100

Q58. What is hyperparameter tuning?

a) Training the model

b) Splitting dataset

c) Optimizing learning parameters like learning rate

d) Removing outliers

59 / 100

Q59. ROC curve stands for:

a) Rate of Classification

b) Receiver Operating Characteristic

c) Regression Optimization Curve

d) Recall of Classifier

60 / 100

Q60. What does the area under the ROC curve (AUC) indicate?

a) Model training time

b) Model accuracy

c) Classification performance

d) Data imbalance

61 / 100

Q61. Which Python library is primarily used for data manipulation and analysis?

a) NumPy

b) pandas

c) TensorFlow

d) matplotlib

62 / 100

Q62. What is the primary use of NumPy in AI development?

a) Plotting graphs

b) Statistical modeling

c) Array manipulation and numerical operations

d) Text processing

63 / 100

Q63. Which Python library provides functions for advanced machine learning tasks and includes tools like GridSearchCV?

a) Keras

b) pandas

c) scikit-learn

d) matplotlib

64 / 100

Q64. Which library is best suited for building deep learning models in Python?

a) seaborn

b) TensorFlow

c) OpenCV

d) nltk

65 / 100

Q65. What does the matplotlib library help you do?

a) Build neural networks

b) Perform matrix multiplication

c) Visualize data

d) Tokenize text

66 / 100

Q66. Which function is used in NumPy to create an array of zeros?

a) numpy.ones()

b) numpy.zeros()

c) numpy.empty()

d) numpy.array()

67 / 100

Q67. Which library works closely with NumPy and provides labeled data structures like DataFrame?

a) TensorFlow

b) matplotlib

c) pandas

d) seaborn

68 / 100

Q68. Which Python package helps in visualizing statistical relationships using beautiful plots?

a) seaborn

b) keras

c) OpenCV

d) scipy

69 / 100

Q69. Which of the following is a feature of the Jupyter Notebook?

a) Compiling Java code

b) Running shell scripts only

c) Interactive code execution

d) Web server deployment

70 / 100

Q70. Which deep learning library is developed by Facebook and used in many AI research projects?

a) TensorFlow

b) Theano

c) PyTorch

d) scikit-learn

71 / 100

Q71. What is the purpose of data normalization?

a) To remove duplicates

b) To reduce memory usage

c) To scale data to a standard range

d) To convert categorical variables

72 / 100

Q72. Which technique helps handle missing data effectively?

a) Label Encoding

b) Imputation

c) Tokenization

d) Regularization

73 / 100

Q73. One-hot encoding is used for:

a) Numerical data

b) Categorical variables

c) Feature scaling

d) Dimensionality reduction

74 / 100

Q74. What is the main disadvantage of one-hot encoding?

a) It can reduce accuracy

b) It adds bias to the model

c) It increases dimensionality

d) It converts data to strings

75 / 100

Q75. Which method is used to detect outliers in data?

a) K-Means Clustering

b) PCA

c) Z-score

d) Min-Max Scaling

76 / 100

Q76. PCA is mainly used for:

a) Data cleaning

b) Feature selection

c) Feature extraction

d) Data balancing

77 / 100

Q77. Which method helps to convert skewed data into normal distribution?

a) Min-max scaling

b) Log transformation

c) Label encoding

d) Cross-validation

78 / 100

Q78. What is binning in data preprocessing?

a) Removing noise

b) Smoothing data by grouping values

c) Deleting null values

d) Transforming features

79 / 100

Q79. Feature scaling is essential for which type of algorithm?

a) Decision Tree

b) Naive Bayes

c) K-Nearest Neighbors

d) Random Forest

80 / 100

Q80. What does L2 normalization focus on?

a) Sum of values

b) Median of values

c) Euclidean norm

d) Maximum value

81 / 100

Q81. What is the primary concern addressed by AI ethics?

a) Hardware limitations

b) Algorithm speed

c) Moral implications of AI actions

d) Software licensing

82 / 100

Q82. What does the term "bias in AI" refer to?

a) AI’s ability to choose between algorithms

b) Human favoritism toward AI

c) Systematic unfairness in AI decisions

d) Lack of data in AI training

83 / 100

Q83. Which of the following is an example of algorithmic transparency?

a) Black-box decision models

b) Explaining how an AI reached its decision

c) Data encryption techniques

d) Speed optimization

84 / 100

Q84. Which principle promotes that AI systems should not cause harm to people?

a) Fairness

b) Safety

c) Accountability

d) Non-maleficence

85 / 100

Q85. What is a major risk of unregulated AI deployment?

a) Higher computational cost

b) Global internet lag

c) Unintended societal impact

d) Reduced battery life

86 / 100

Q86. What is explainable AI (XAI) designed to address?

a) Improving the training speed

b) Reducing data size

c) Interpretability of AI decisions

d) Data augmentation

87 / 100

Q87. Which body often provides guidelines for ethical AI in Europe?

a) NASA

b) European Union

c) World Bank

d) IEEE

88 / 100

Q88. What role does accountability play in responsible AI?

a) It defines data formats

b) It traces AI actions to responsible individuals or entities

c) It encrypts model weights

d) It ensures fast computation

89 / 100

Q89. A hiring AI that rejects applicants from a certain race shows a failure in:

a) Efficiency

b) Reliability

c) Fairness

d) Bandwidth

90 / 100

Q90. What is the best practice to mitigate bias in AI models?

a) Using only image data

b) Collecting diverse and representative datasets

c) Training on synthetic data

d) Ignoring edge cases

91 / 100

Q91. Which company is known for using AI to optimize delivery routes through its "Dynamic Routing" system?

a) FedEx

b) Uber

c) Amazon

d) Walmart

92 / 100

Q92. In healthcare, IBM Watson was initially developed to assist with:

a) Drug design

b) Diagnosis and treatment recommendations

c) Robot-assisted surgery

d) Fitness tracking

93 / 100

Q93. Which AI application is commonly used in credit card fraud detection?

a) Neural style transfer

b) Generative adversarial networks

c) Pattern recognition

d) Sentiment analysis

94 / 100

Q94. Google’s AlphaFold AI made a breakthrough in which scientific area?

a) Quantum computing

b) Astronomy

c) Protein structure prediction

d) Robotics

95 / 100

Q95. Which AI model is used by YouTube to recommend videos?

a) Naive Bayes

b) Reinforcement Learning

c) CNN

d) Recommendation System based on Deep Learning

96 / 100

Q96. Tesla’s self-driving feature relies heavily on:

a) Natural language processing

b) Computer vision and sensor fusion

c) Speech recognition

d) Predictive analytics

97 / 100

Q97. Which AI technology is behind Apple’s Siri?

a) Rule-based system

b) Reinforcement learning

c) NLP with speech recognition

d) Computer vision

98 / 100

Q98. In e-commerce, which AI technique helps suggest products to users?

a) Clustering

b) Recommendation engines

c) GANs

d) Anomaly detection

99 / 100

Q99. Which company implemented AI to reduce customer churn using predictive analytics?

a) Netflix

b) IBM

c) Coca-Cola

d) HSBC

100 / 100

Q100. AI-powered facial recognition systems are predominantly used in:

a) Agriculture

b) Transportation

c) Surveillance and security

d) Food processing

Your score is

The average score is 75%

Exit

What You Learn in an AI Internship

Leave a Reply Cancel reply