Umamaheswararao B.

Umamaheswararao B.

Data Warehouse Consultant

Vijayawada , India

Experience: Year

Umamaheswararao

Vijayawada , India

Data Warehouse Consultant

18683.8 USD / Year

  • Immediate: Available

Year

Now you can Instantly Chat with Umamaheswararao!

About Me

  • Having 3+ years of Experience in developing scripts using Python, Machine Learning.
  • Experienced on Data Types, functions, OOP, File Handling, Exception Handling, packages and modules concepts usin...
  • Skilled in libraries such as NumPy, Pandas, Matplotlib, Sklearn (Scikit-learn), SciPy, Seaborn, Tableau for Data Visualization, Keras, TensorFlow.
  • Experience in architecting applications with Machine Learning algorithms like Linear regression, logistic regression, Decision Tree, Naïve bayes classifier, SVM, KNN, Random Forest, K-means 
  • Hands on Experience on Databases such as SQL and MySQL.
  • Good experience on DBMS concepts Like SQL, Normalization and ER Diagrams.
  • Knowledge on Data Warehousing and Data Mining such as preprocessing, Association, Classification and clustering Techniques.
  • Knowledge on cloud computing technologies such as Sales force, AWS.
  • Hands on experienced on GIT repository, PyCharm IDE.

Show More

Portfolio Projects

Description

The scope of this project is to predict the property price based on certain features like square feet, bed rooms, bath rooms and locations etc. Take home price data set from company. Using data set and machine learning techniques to build model. Cover the Data science concepts like Data cleaning, Feature Engineering, Dimensionality reduction, outlier detection, model development and deployment in flask servers.

  • Participating in Data Preprocessing Techniques in order to make data useful for creating Machine Learning Models
  • Building various regression and classification algorithms by using various Sklearn libraries such as Linear Regression, Decision Trees and Naive Bayes
  • Develop and maintain our Python platform codebase using Python scientific packages such as NumPy, Pandas, sklearn, Matplotlib and Python Flask.
  • Help enforce software best practices, and work with others to deploy new models to our production servers
  • Work with an interdisciplinary team and drive projects to completion
  • Strong skills in data manipulation, cleaning and exploratory analysis. Able to make logical conclusions about complex data that help properly format it for statistical analysis
  • Analyze on system or data processes, data structures, analytical methods, and applying key techniques of data sciences (machine learning, data mining, analysis) to build high quality predictive system.

Show More Show Less

Description

A large number of documents produced in an enterprise are in the form of scanned images. Tables within PDF or scanned documents convey relationship between multi-dimensional data elements and attributes in a compact (high information density) and easily readable format. For example, in an invoice document, the line items are placed in a tabular from which conveys the number of items purchased, amount etc. OCR tools are useful to extracting the data from document images, however, they possess very limited capability to extract the desired data from tables. Table data extraction is complex and challenging task due to its structural properties and other constraints (such as breaking a column value into two or more lines. The problem of table extraction can be divided into boundary detection and then decomposition of the text. This project aims at retrieving text that is represented in the form of table from document images using machine learning techniques and domain-specific heuristics. It also exploits the structural properties and other information relating to the text region within the document to extract accurate data from tables present in the document images.

  • Work with the business to apply scientific methods and to evolve data science solutions
  • Exposure to SQL and databases
  • Work with an interdisciplinary team and drive projects to completion
  • Strong skills in data manipulation, cleaning and exploratory analysis. Able to make logical conclusions about complex data that help properly format it for statistical analysis
  • Analyze on system or data processes, data structures, analytical methods, and applying key techniques of data sciences (machine learning, data mining, analysis) to build high quality predictive systems
  • Visualization of complex data and analytical result using modern data visualization tool
  • Identify hidden pattern or systematic risks using qualitative and quantitative data analysis
  • Develop solution allowing highly effective presentation and engaging visualization for decision making
  • Work with the data scientists to develop a Python codebase for predictive and prescriptive modeling, along with the development of other data science capabilities
  • Develop data solutions using multiple sources, such as SQL, Google Analytics, and other platform API’s

Show More Show Less

Description

Computer based bio-electronic systems are used for replacement of damaged human parts such as Bionic-ear for deafness, Bionic-eye for blindness, Deep Brain Stimulator for diseases of the brain, and Bionic-arm for arm prostheses. Algorithms for controlling bionic system are based on the specific bionic devices like bionic ear with sound processing software and bionic eye with image processing software. As the use of software in medical devices has improved, the need for specific regulations for healthcare system software has improved. Software for medical systems has to deal with the Health software lifecycle identified by safety analysis in order to make the system safe. This project aims at retrieving text that is represented in the form of table from document images using machine learning techniques and domain-specific heuristics. It also exploits the structural properties and other information relating to the text region within the document to extract accurate data from tables present in the document images.

Show More Show Less