Artificial intelligence has different subfields and machine learning is one of those subgroups. Machine learning is basically the machine’s capability to emulate intelligent human behavior. Graph machine learning is a plate form which gives new set of tools for processing network data and capitalizing the power of connection between structures that can used for the predictive, modeling and analytical tasks. These notes are very useful for the machine learning developers who wants to make machine learning driven graph databases. You will learn all concepts of graph theory, techniques and algorithms which is used to make successful l learning application.

**In these notes you will learn following things**

First you will learn about graph representation method and will know how to visualize graphs and the types of graphs. You will know the properties which are used to characterize network. Then you will learn some examples which are generally used to study the network’s properties, effectiveness of networks algorithm and benchmark performance and you will learn about the dealing of large graphs. Next you will learn about the basic concepts of the machine learning and the way to apply them on the graphs. There is information about the technical requirements used in understanding machine learning on graphs, basic principles of machine learning and the benefits of machine learning on graphs. You will learn about the generalized graph embedding problems and the taxonomy of graph embedding machine learning algorithm. There are examples which are provided to understand how the theory can be applied to practical problems.

These notes will provide you knowledge about unsupervised graph learning that how it can be applied to the graphs to solve the problems effectively. You will learn about technical requirements, unsupervised graph embedding roadmap and shallow embedding methods which is a set of algorithms used to learn and return only embedding values. It provides information about autoencoders which is used to encode any input without damaging the important information. You will also learn about graph autoencoders. Then you will learn about graph neural networks (GNN), the concepts of GNN, variants of GNN, spectral graph convolution, spatial graph convolution and graph convolution. Next you will learn about how to solve real problems by applying machine learning. You will learn about how to train classic machine learning algorithms by using graphs and nodes as features directly. It gives information about graph regularization method and how it can be used to create more robust models during the learning phase that tends to generalized better. Then you can see that how supervised machine learning problems can be solved by applying GNNs.

Next you will learn about different problems with machine learning on graphs and how it can be solved by graph based machine learning techniques. You will see that to solve apparently different tasks same algorithm can be adapted. Each problem has its own peculiarities. You will learn about detecting graph similarities and graph matching GNN based methods and applications. Then you will learn about social network graphs and how on this for solving problems machine learning is useful. Further you will learn that how SNAP face book combined ego network can be predicted by future connections. You will learn about network topology and community detection, embedding for supervised and unsupervised tasks. Then you will learn about text analytics and natural language processing using graphs and will learn how to represent unstructured information by graphs. You will understand the main concepts and tools used in NLP, will learn to create graphs from a corpus of documents and you will learn how to build a topic classifier.

Next you will learn about graph analysis for credit card transaction. You will learn that in order to spot and plot specific regions of the transaction graph a community detection algorithm was applied. You will learn about the series of examples that can be used to show how it is applied to problems related to different domains. Then you will learn about building a data driven graph powered applications in which you will learn about the basic concepts of designing, implementation and deploy data driven applications. You will learn about lambda architectures for graph powered applications, graph processing engines, graph querying layers and selection between Neo4j and graphX. In last you will learn about novel trends on graphs. In this you will learn about exploring data augmentation techniques, topological machine learning, graph machine learning and neuroscience, graph theory and chemistry and biology and graph machine learning and computer vision. You can download these notes totally free.