Sanjiv Anto

I worked on a number of automation projects from the initial planning stage of the control system all the way through to commissioning and client handover. I worked with different control systems in 22 different countries to automate plants that prepare seeds, extract oil, and refine edible oil.

The goal of this project is to deliver an analysis of a Moroccan power consumption dataset and the application of a predictive model using machine learning techniques. The Linear Regression machine learning algorithm was used to predict power consumption based on the measured atmospheric temperature. The entire project was completed on the Google Cloud Platform with Google Cloud Storage, Dataprep, and Big Query.

In this project, I created tree-based machine learning models to predict whether or not a person has the disease or not using the “diabetes.csv” dataset provided.

I programmed and evaluated tree-based classification models using multiple learning algorithms including Decision Trees, Random Forests, AdaBoost and Gradient Boosting.

Business Scenario: An engineering firm specialized in supplying edible oil refinery plants and
equipment intends to apply data science on one of its refinery installations to investigate the
potential for energy savings and throughput enhancements, which will help them improve their
engineering and design for future projects.

I used exploratory data analysis to investigate the automation dataset in order to discover previously unseen trends in the data, which can assist organizations gain insights from this data.

The Cloud Pub/Sub to BigQuery DataFlow template—a powerful tool within the Google Cloud ecosystem—is the focus of this project. This was done in order to figure out how to set up a streaming data pipeline that receives JSON messages from taxi drivers, analyzes them, and publishes the data to a BigQuery table. BigQuery Geo Viz was used to display taxi riders’ locations.

In this project, I did an image classification model using convolution neural networks (CNNs) and the cats-and-dogs images dataset from Kaggle.

I did the following in this project:

1. Download the dataset
2. Explore and preprocess the data
3. Preprocess and prepare the data for modeling
4. Develop a baseline model using 1, 2, and 3 blocks as in the VGG model
5. Train and evaluate the models and report their results

This was a case study condusted in order to understand about LOTTE’s remarkable journey of migrating their S/4HANA operating infrastructure from on-premises to Google Cloud. As a company with over two decades of experience running SAP, LOTTE recognized the need for a zero-base core system renewal as the support for their current ERP solution approached its end of life. In this case study, we will delve into the details of LOTTE, their selection of SAP’s Managed Services, and their subsequent decision to migrate to Google Cloud, showcasing the significant benefits achieved

The primary objective of this project was to enhance my understanding of the linear regression machine learning model. The project was executed using Jupyter Notebook, a popular interactive development environment, with Python as the programming language. To implement the linear regression model, I leveraged the Scikit-Learn package, a robust machine learning library. The completion of this project enabled me to gain valuable insights into the workings of linear regression, its applications, and its practical implementation using these tools.

The goal of this project was to train and test classification models with various algorithms and then compare their performance on the test dataset using various measures of performance.

Introduction:
In this project, I worked with a real and open dataset, the Chicago Crime
Dataset. This dataset reflects reported incidents of crime (with the exception of
murders where data exists for each victim) that occurred in the City of Chicago. Data
is extracted from the Chicago Police Department’s CLEAR (Citizen Law Enforcement Analysis and Reporting) system. In order to protect the privacy of crime victims, addresses are shown at the block level only and specific locations are not identified.

Objectives:
I tried to to explore, analyze, and draw inferences/suggestions with a
comprehensive report about the crime to the government of the City of Chicago.

Retrieve data from the employee database and provide information that will enable management to take a more data-driven approach to decision-making. In this project, I used SQL query language to achieve the necessary results.