Aspire Journeys
Predictive Analytics Journey
- 17 Courses | 22h 15m 52s
- 1 Lab | 4h
While data is one of the most valuable assets of an organization, predictive analytics has become the most important task to examine this data, using business knowledge and extracting valuable insights and interesting patterns. Organizations use predictive analytics for various advantages including understanding their customer base, improving their operations, increasing revenues, outperforming their competitors, and better positioning themselves in the marketplace.
Track 1: Predictive Analytics
In this track of the Predictive Analytics Skillsoft Aspire journey, the focus will be on the applications, use cases and research in 5 major domains such as Cyber security, Operations, Marketing and Retail, Agriculture, and Healthcare. Along with various interesting scenarios in each domain where predictive analytics is applied, the focus is also to go through the interesting case studies with hands-on python code and applied machine learning algorithms.
- 17 Courses | 22h 15m 52s
- 1 Lab | 4h
COURSES INCLUDED
Predictive Analytics: Case Studies for Cybersecurity
Cybersecurity is the protection of user software from maliciously-intentioned agents and parties. Cyberattacks commonly focus on critical physical infrastructures like power plants, oil refineries, and gas pipelines. For geopolitical reasons, the cybersecurity of such installations is increasingly important. In this course, explore the use of classification models when modeling cyberattacks and the evaluation metrics for classification models. Next, examine a case study where machine learning and cybersecurity attempt to detect intrusions in a gas pipeline. Finally, investigate a case study where machine learning models are used to detect and cope with malware. Upon completion, you'll be able to identify the need for AI in cybersecurity and outline the appropriate use of evaluation metrics for classification models.
10 videos |
1h 23m
Assessment
Badge
Predictive Analytics: Identifying Network Attacks
In cybersecurity, it's important to determine whether a user interaction or action represents an attack, followed by discerning the specific attack type and signature. Machine learning (ML) models and managed ML solutions like Microsoft Azure Machine Learning can help with this. In this course, learn how to create an Azure Machine Learning workspace, read in data, and categorize all of the different types of attacks. Next, discover how to train a random forest classification model using the scikit-learn library and test it on the in-sample validation data. Finally, practice performing multiclass classification to identify the specific type of attack. Upon completion, you'll be able to detect intrusions using data, train and evaluate classification models, and perform multiclass classification.
18 videos |
2h 3m
Assessment
Badge
Predictive Analytics: Case Studies for Operations
Nowadays, the domains of operations management and supply chain management are growing ever more complex. Operations management is the management of labor, resources, and processes for goods manufacturing for customers. Supply chain management encompasses the timely procurement of materials for factories and the safe and timely transportation of finished products to the end customer. In this course, explore a case study about leveraging machine learning models for identifying reliable suppliers. Next, examine a study that attempts to identify failures of induction motors using machine learning models. Finally, discover how analytics can be used to predict order lead time and demand in an aluminum firm. Upon completion, you'll be able to outline use cases for the application of AI to operations and supply chain management.
11 videos |
1h 36m
Assessment
Badge
Predictive Analytics: Identifying Machine Failures
Machines are the building blocks of manufacturing facilities, and facilities have grown larger and more complex over time, resulting in the increased effects of a single machine failure. Artificial intelligence (AI) and machine learning (ML) models are commonly used to quickly get ahead of and diagnose machine failures. In this course, learn how to create an Azure Machine Learning workspace for building models that predict machine failures. Next, practice performing preprocessing tasks on machine failure prediction data. Finally, discover how to perform machine failure prediction using logistic regression. Upon completion, you'll be able to detect machine failures using machine learning methods.
11 videos |
1h 10m
Assessment
Badge
Predictive Analytics: Using SMOTE, Model Explanations, & Hyperparameter Tuning
Machine learning (ML) models can struggle with training themselves to identify failures if the dataset's number of machine failures is too low. This is a common problem that occurs when predicting very rare occurrences. Thankfully, oversampling techniques exist to mitigate such issues. In this course, learn how to use SMOTE, a widely used technique to make datasets more balanced. Next, explore model explanations, a feature of Azure Machine Learning. Finally, practice performing hyperparameter tuning by trying different model configurations to see which yields the best performance. Upon completion, you'll be able to improve the performance of a failure detection model, generate records of minority classes, and perform hyperparameter tuning.
11 videos |
1h 15m
Assessment
Badge
Predictive Analytics: Case Studies for Marketing & Retail
Artificial intelligence (AI) can play a part in many marketing tasks, and if used skillfully, can have a significant impact on revenues and profits. Every action on social media is recorded and AI is well-suited to process this large volume of data. AI can also be used to help with demand forecasting, product recommendations, and fraud detection. In this course, you will investigate significant ways to apply AI in marketing and in retail. Explore the Weibo case study, which focuses on data from marketing and advertising posts on social media and the attempts to predict likes, comments, and shares of each post before it is sent out. Then, analyze the applications of market basket analysis (MBA) via case study to determine how to perform customer segmentation using clustering and MBA through association rules mining. Finally, discover how to predict out-of-stock events in a store.
9 videos |
55m
Assessment
Badge
Predictive Analytics: Predicting Sales & Customer Lifetime Value
In recent years, retailing has changed from a fragmented space into a winner-takes-all sector, in which a key differentiating factor is the ability to tightly predict demand and measure customer lifetime value. Begin this course by attempting to predict the sales for each week in a Walmart store. You will explore and visualize your data, creating an Azure machine learning workspace and a hosted Python notebook to write code. Then, perform regression analysis to predict the sales after one-hot encoding the requisite explanatory variables. You will apply different models as well, including ridge regression, K-nearest neighbors, decision trees, random forests, and extra tree regressors. Next, predict the customer lifetime value using regression analysis, and perform cross-validation and feature selection on the model in order to improve its performance. Finally, experiment with feature selection, including recursive feature elimination, lasso regularization, and linear SVR.
14 videos |
1h 41m
Assessment
Badge
Predictive Analytics: Predicting Responses to Marketing Campaigns
While growth has become a primary driver of business valuations and key performance indicators for executives, a renewed emphasis on profitability has led to greater attention being placed on cost-effective, tailored strategies for customer acquisition. This makes predicting the likely response of a prospect to a marketing campaign an increasingly important skill. Discover how to predict the purchase intention of shoppers on e-commerce websites, beginning with cleaning the data and then exploring it for patterns using heatmaps and correlation analysis. Next, perform classification analysis to predict whether a shopper will purchase something. Then, explore different evaluation metrics such as the confusion matrix, recall, precision, and the F1-score. Finally, you will predict whether a specific user responded to a marketing campaign and use feature selection to optimize the number of explanatory variables used in the process.
13 videos |
1h 16m
Assessment
Badge
Predictive Analytics: Customer Segmentation & Market Basket Analysis
Two central goals of marketing are reducing customer acquisition costs and increasing customer lifetime value. Customer segmentation is an important step towards both of these goals - by learning more about present and prospective customers, marketing practitioners can focus on tailoring strategies to acquire and retain these different types of customers more effectively. Explore the regency, frequency, and monetary value (RFM) framework of customer interactions by performing K-means clustering and using the silhouette score to pick the optimal number of clusters. Next, switch to two alternative clustering techniques, known as agglomerative clustering and DBScan. Finally, perform market basket analysis, also known as affinity analysis, to predict what items that customers will purchase together, such as bread and jam. Use the a priori algorithm for computing frequent itemsets, and the calculation and implications of metrics such as support, confidence, lift, and conviction.
16 videos |
1h 52m
Assessment
Badge
Predictive Analytics: Case Studies for AI in Agriculture
Population growth, climate change, and volatile commodity prices risk factors are putting a strain on the agricultural system nowadays. Using artificial intelligence (AI) in agriculture can potentially help mitigate this strain in areas such as yield prediction and disease detection in agriculture. In this course, explore a study that uses machine learning (ML) models for agricultural use cases. Next, explore a specific case study that attempts to predict the yield of maize and soybean crops on various American farms. Finally, examine a study that uses machine learning for pest detection. Upon completion, you'll be able to gather and analyze academic papers on machine learning in agriculture, identify problem categories and solution constructs, and recall common recurring themes in research.
10 videos |
58m
Assessment
Badge
Predictive Analytics: Performing Classification Using Machine Learning
Species identification is an important step in many agricultural processes. With supply chain globalization and strict grading norms imposed by importing nations on the composition of consignments, machine learning (ML) techniques geared towards classification are becoming increasingly relevant in agriculture. In this course, learn how to perform exploratory data analysis on data about beans. Next, discover how to visualize the data to get a sense of the relationships between the attributes, perform logistic regression to classify the beans, and explore the result metrics of the model. Finally, practice performing feature selection and using other types of machine learning models. Upon completion, you'll be able to analyze agricultural data, recognize relationships between attributes, and classify data based on type.
13 videos |
1h 21m
Assessment
Badge
Predictive Analytics: Applying Clustering to Soil Features & Conditions
The question of what crops ought to be planted per growing season for a given patch of land is extremely important. An important related question is which type of crop fits most easily with the soil and climatic conditions. Machine learning (ML) models like clustering can help answer this question using data from other farms. In this course, work with soil data consisting of field climate conditions. Next, learn how to use charts to view univariate information and the relationships between attributes. Finally, discover how to perform k-means and agglomerative clustering on data. Upon completion, you'll be able to apply clustering to data, identify links between clusters identified by ML algorithms and the crops cultivated in them, and differentiate k-means and agglomerative clustering.
12 videos |
1h 24m
Assessment
Badge
Predictive Analytics: Performing Prediction Using Regression
In agriculture, accurately assessing crop yield in advance can help farmers effectively plan ahead, allocate labor and capital, and plan for crop transportation logistics. Machine learning (ML) can be used to account for the many factors that drive yields. In this course, work with data consisting of blueberry plant information and climate factors to predict yield. Next, learn how to visualize univariate relationships and bivariate correlations and perform linear regression. Finally, practice performing feature selection for the regression model and view the score of importance and model on a subset for different data attributes. Upon completion, you'll be able to use regression techniques to predict agricultural yields, identify real-world and statistical relationships in the data, and differentiate between various regression models.
8 videos |
54m
Assessment
Badge
Predictive Analytics: Case Studies on Predictive Analytics for Healthcare
Healthcare aims to improve the health of individuals, but generally, healthcare systems tend to be extremely strained. Using artificial intelligence (AI) with healthcare could potentially mitigate this strain on the system. In this course, examine how AI is used in healthcare, how to evaluate classification models, and the metrics that are significant in models used in disease diagnosis. Next, discover the importance of a model's recall or sensitivity and the computation of the ROC curve and AUC metrics. Finally, explore the process of compiling the datasets, training, and evaluating models from research papers that take a general look at the application of AI in disease and specific ailment diagnosis. Upon completion, you'll be able to identify healthcare use cases for AI and its limitations.
11 videos |
1h 27m
Assessment
Badge
Predictive Analytics: Detecting Kidney Disease Using AI
Nowadays, diseases such as Alzheimer's, heart disease, and diabetes are becoming ever more prevalent across the world. Use this course to get hands-on experience building a pipeline to diagnose chronic kidney disease using Azure Machine Learning designer. Explore the different features of Azure Machine Learning, its interface, and how components and resources come together to build a pipeline. Next, learn how to build a pipeline to create a dataset, implement various data cleaning tasks, and work with the cleaned dataset to build a logistic regression model to detect kidney disease. Finally, examine how models can be trained and evaluated for performance and deploy your pipeline. Upon completion, you'll be able to build and deploy a disease diagnosis Azure Machine Learning pipeline.
17 videos |
1h 47m
Assessment
Badge
Predictive Analytics: Identifying Tumors with Deep Learning Models
Azure Machine Learning designer allows you to create machine learning models using no-code, drag-and-drop pipelines. Use this course to build pre-trained neural network models that detect diseases from image scans using Azure Machine Learning designer. Learn how to set up data for model training, validation, and testing and how to feed that data into a pipeline that employs a DenseNet model. Next, discover how a model can be configured and substitute a pipeline's DenseNet model with a ResNet model. Finally, explore how a model's training metrics can be analyzed to understand what tweaks need to be applied to build a more reliable model. Upon completion, you'll be able to build DenseNet and ResNet models that can identify tumors from chest scan images.
10 videos |
1h 4m
Assessment
Badge
Final Exam: Predictive Analytics
Final Exam: Predictive Analytics will test your knowledge and application of the topics presented throughout the Predictive Analytics journey.
1 video |
32s
Assessment
Badge
EARN A DIGITAL BADGE WHEN YOU COMPLETE THESE TRACKS
Skillsoft is providing you the opportunity to earn a digital badge upon successful completion on some of our courses, which can be shared on any social network or business platform.
Digital badges are yours to keep, forever.
