Data Analytics

We have Data Scientists who have done masters in Machine Learning and have working knowledge with the real data. We have already taken a step into various data handling aspects like Data cleaning ,Exploratory analysis , Regression , Classification ,Recommender systems,Network analysis , Predictive analytics and good working knowledge of visualization of the same. We use R, Java, Python, Hadoop,Spark, PMML , ggplot2 , D3,Pentaho Kettle ,Angular JS ,RestJ and a bit more.

Our trek to the Data Analytics Summit

Understanding the actual business objective and what we have to achieve that (Data)

Data understanding and Data cleaning

Getting hold of data with help of tools like Kettle, Talend, Hadoop, Casandara, Spark for transformation and processing. As in any other data analytics project , this is the most important step in our data analytics life cycle too. Also in this phase , the quality of the data is to be ensured good.

Exploratory analysis , Feature extraction and Data modeling

This phase can boast of being the most challenging yet exciting phase of the data analytics life cycle. In this phase we have to identify which features or what combination of features can be used for the modeling based on the information gain it has.We have to check for the frequency distribution, correlation of the features and many other statistical methods.

Based on the business challenge and data in hand , we can approach the problem through the statistics and machine learning algorithms . The tools used in this phase are standard technologies like R , Python. This can used in the java world by converting the model into PMML .

Data Visualization

The exploratory analysis and the results make sense to the customer more , as they will be able to catch the latent information hidden in the data. And hence this phase has a significant role in the decision making process as it makes the interpretation of the data easy. We use Angular JS, D3, RestJ for integrating visualizations to the system.

The last leap

The last phase is to fine tune the model and do custom analytics as per the need of the customer. This can be an iterative process.

Moral of the Story

If we can climb a data analytics mountain with the tools and expertize in our kit , so can we climb the other data analytics mountains and help our customer reach the top.