Professional Certificate in Machine Learning

Explore how Python uses variables as boxes, assign values, and perform arithmetic. See how x and y hold values, update with new numbers, and view outputs reflecting changes.

Explore variables in Python and string operations, including concatenation, zero-based indexing, and negative indexing, with examples like names and social networks; learn how index errors occur when out of range.

Explore Python variables and string operations, using zero-based indexing and slicing with examples like Facebook or name strings, and learn how to get string length.

Practice Python for absolute beginners by creating lists, storing multiple values, and printing them to observe variables, different types, and integers.

Explore lists in Python by creating lists of numbers and strings, appending values at the end, inserting at specific positions, removing items, and using zero-based indexing.

This lecture continues exploration of python lists, showing how to perform operations like indexing, popping an element, and checking membership, with examples of min, max, sum, mean, and sort.

Identify assumptions, discard irrelevant information, and clearly define the problem before solving. Learn to express solutions as algorithms using flowcharts and a simple input–process–output model.

Explore flowcharts as a graphical representation of processes. Learn standard symbols for start/end, input/output, and decision making, and apply sequence, selection, and iteration.

Explore repetition in programming, detailing the three core structures: sequence, selection, and repetition, and compare while loops, repeat-until, and do-while styles, including pre- and post-condition evaluation.

Explore problem solving with flowcharts, using standard symbols for input, processing, output, and start/stop to illustrate algorithms like summing numbers, temperature conversion, and area calculations.

Explore how statistics help understand data for machine learning by inspecting raw data, reading from a file, and listing the first 50 records to reveal data quality and structure.

Explore how to check data dimensions, rows and columns, assess size, and obtain mean, standard deviation, min, max, percentiles, and class distribution for classification readiness.

Explore how statistics measure relationships between attributes with a correlation coefficient, including full positive, full negative, and zero correlations, and how this affects linear and logistic regression.

Explore how correlation measures the relationship between two variables using the Pearson correlation coefficient, including positive, negative, and zero correlations, and why high correlation impacts linear and logistic regression.

Learn how to prepare data for machine learning by preprocessing, scaling data from 0 to 1 with normalization in Python, and ensuring data integrity and privacy for reliable models.

Explore data preprocessing techniques in Python, mastering normalization, binarization with a threshold, and standardization to a Gaussian distribution, and see how these steps support linear and logistic regression.

Learn how univariate feature selection identifies the most relevant data features using statistical tests to strengthen the relationship with the target prediction, boost accuracy, and reduce training time.

Learn to prepare data and visualize it for machine learning by loading datasets, building bar charts in a Python notebook, and interpreting category insights.

Learn to visualize data with Python using histograms and pie charts, analyze the distribution and skewness of numeric data across continuous intervals, with food items and cuisines as examples.

Explore how artificial neural networks mimic brain processes to solve problems with machine learning, and examine the single-input neuron model, weights, bias, net input, and activation functions.

Explore artificial neural networks from scratch by implementing a single neuron in Python, summing weighted inputs x1, x2, x3 with a bias, before applying any activation function.

Explore how multiple input neurons combine several inputs via weights and biases to produce outputs using a transfer function. Contrast this with single input neurons and related network architectures.

Learn how to build a simple neural network from scratch in Python, with an input layer, a four-neuron hidden layer and an output layer, using weights, biases, and activation functions.

Explore an illustrative neural network example using a perceptron to classify apples and oranges from three inputs: shape, texture, and weight, based on learned weights and bias.

Develop a Python Keras neural network from a diabetes dataset, using eight inputs, sequential model with relu hidden units and sigmoid output, trained with binary crossentropy and stochastic gradient descent.

Develop a step-by-step deep learning workflow to recognize handwritten digits from 28x28 images, using a sequential model, dense layers, relu, softmax, normalization, and accuracy-driven training.

Explore the naive bayes classifier through a hands-on demo that uses probabilistic reasoning and independent feature contributions to classify data, with multinomial and gaussian approaches and spam vs. not spam.

Set up your machine learning environment by downloading and installing Anaconda, using the graphical installer, and launching Jupyter Notebook via Anaconda Navigator to run ML examples.

Explore tokenization in natural language processing, learning how to split text into words and sentences, handle punctuation, apostrophes, hyphens, and other challenges using a Kopra dataset and NLP tools.

Install and set up the natural language toolkit (NLTK) with Python, configure your IDE, install packages and models, and begin with tokenization using the analytical modules.

Explore tokenization in natural language processing by tokenizing text into sentences and words in Python, printing tokens, and preparing data for preprocessing.

Explore normalization in preprocessing by converting words to a uniform form to improve information retrieval, and examine tokenization, stemming, lemmatization, and handling prefixes, suffixes, and capitalization.

Learn how stemming and lemmatization normalize word variants in natural language processing using Analytica, with examples like illegal to legal and legalize to illustrate simpler corpus processing.

Explore how part of speech tagging labels each word in a sentence to aid preprocessing in natural language processing, with tokenization and tag outputs illustrated.

Learn how part of speech tagging labels each word with its part of speech as a preprocessing step, using tokenization and example sentences to illustrate the process.

Identify stopwords as noise in natural language processing and learn how filtering them from tokenized sentences reduces processing load, paving the way to named entity recognition.

Explore how named entity recognition identifies and classifies names in text, such as persons, organizations, and places, and enables indexing and question-answering systems.

Practice named entity recognition using tokenization, tagging, and chunking to identify entities in text with analytics. Learn how person, organization, GPE, and places are detected, plus common misidentifications.

Explore how text classification uses the naive algorithm to label emails as spam, analyze sentiment in reviews, and identify topics, languages, and authorship.

Explore text classification and sentiment analysis by preprocessing movie reviews using Naive Bayes. Shuffle the corpus, lowercase words, and create document-word tuples to identify the most common features.

Apply text classification using natural language processing by identifying the top 3000 words via frequency distribution, building word features, and training a positive and negative review classifier.

Finish classification with the naïve Bayes method using the Analytica library, a 75/25 train-test split, and training on 1500 features. Achieve 82% accuracy and identify features like outstanding and idiotic.

Experiment with a sentiment analysis classifier in natural language processing by tweaking training/test splits, adjusting most common words, applying normalization, and removing stop words to boost accuracy beyond 82 percent.

Explore a real world natural language processing project by building a Twitter sentiment analysis tool that classifies tweets as positive or negative, using a 5000-tweet CSV training dataset.

Create a Twitter application to obtain keys and tokens, configure the API, and verify credentials for sentiment analysis with natural language processing.

Fetches the 100 most recent tweets for a keyword via the Twitter API to build a test set for a real-world Twitter sentiment analyzer, preparing data for labeling.

Prepare the training set by converting the 5000 tweet corpus into a text and sentiment dataset, fetching tweet text by ID, and writing the final training file.

Learn to preprocess real-world text for sentiment analysis by building a Python class that lowercases, removes punctuation, substitutes mentions and URLs, and tokenizes tweets, preparing training and test sets.

Build a tweet sentiment classifier by creating a vocabulary from training data, extracting and applying features, training a Nyberg's classifier, and classifying test tweets for overall sentiment with a percentage.

Build a sentiment analyzer from Twitter data, training and test sets, preprocessing tweets, and extracting features to classify sentiment as positive or negative. Apply natural language processing to projects.

Explore Python basics by teaching variables as boxes, assignment with X and Y, and simple operations to update and output values, including results of previous operations.

Learn to work with variables and strings in Python, in part two, using zero-based and negative indexing to retrieve characters and understand index out of range errors.

Learn how to work with variables and strings in Python, including indexing and slicing from zero-based indices, using examples like Facebook, and calculating string length.

Explore python for absolute beginners by learning about variables and lists, printing list contents, and mixing strings and integers, highlighting python's flexible list types unlike java.

Learn to manipulate lists in Python by appending, inserting at specific positions, and removing items, using zero-based indexing with numbers, strings, and mixed lists.

Practice completing list operations in Python, including indexing, popping first and last elements, checking for removal, and computing min, max, mean, sum, and sorting the list.

Compare linear regression and logistic regression using an age-based insurance example to show why linear regression cannot handle non-linearly separable data, and how logistic regression provides a decision boundary.

Explore univariate linear regression with a hands-on demo that models happiness from GDP per capita, and learn data loading, plotting, and an 80/20 training-testing split.

Explore univariate linear regression with GDP per capita to predict happiness, train via gradient descent to minimize the cost function, and evaluate predictions on training and unseen data.

Explore a hands-on multivariate linear regression demo that uses GDP per capita and freedom to predict happiness, with training and test sets, gradient descent, and visualized 3d plots.

Apply logistic regression to classify insurance purchase using age, weight, and status as features. Train and test the model, then tune parameters to improve predicted probabilities.

Group data into natural clusters to reveal the internal structure of data, maximizing intra-cluster similarity; treat clustering as unsupervised learning without labeled data.

specify the number of clusters, randomly initialize centroids, and iteratively assign points to the nearest centroid while updating centroids as the mean of points in each cluster until convergence.

Learn to implement k-means clustering in python with a hands-on, step-by-step approach. Cluster iris flowers into three species using length and width features, then evaluate against known labels.

Practice k-means clustering with a Python code walkthrough. Learn unsupervised learning, centroid initialization, nearest-centroid assignment, and mean-based centroid updates.