Easy Steps to Use Google Big Data Query and Google Prediction

Lets understand the basic of Big data and then how you can use it.

Big Data: This is a large amount of data which can be hosted on Google, it can be from social media, Email, Google search engine, online video repository, various public and private transactional databases etc.

Big Query: it works on Google internal technology called “Dremel”. Dremel allows to do a SQL like query on any massive amount of data set and return the result in few seconds.
According Google infrastructure guru Urs Hölzl “You have a SQL-like language that makes it very easy to formulate ad hoc queries or recurring queries– and you don’t have to do any programming. You just type the query into a command line”

Why to use Google Big Data and Big Query:

1. Store data with reliability, redundancy, and consistency
2. Go from data to meaning
3. At Scale
4. Quickly

Where to use it:

1. Game and Social Media Analytics
2. Infrastructure Monitoring
3. Advertising campaign Optimization
4. Sensor data Analysis

To do it by yourself:
If you want to see the data processing speed, you can do it yourself in a very easy few simple steps

1. Click the link https://bigquery.cloud.google.com/
2. Login using your Gmail user id and password (To access any of the Google
3. product you always require Gmail User id and password).
4. Click compose query button at the top left of the page.

5. Paste following query in the New query box.
SELECT Year YEAR,
AVG(cigarettes_per_day) CIG_PER_DAY,
AVG(gestation_weeks) GES_WEEKS,
FROM (publicdata:samples.natality)
WHERE cigarette_use=true
GROUP BY Year
6. And click run query below the New Query Box.
7. You can see “RUN QUERY” tab the tab taken to complete the query as well as volume of data processed.

The above example which you have seen is based on the sample data, the same can be done with our own data as well.

Google prediction API

Google prediction API works in three steps
1. Structure the training data and Upload to Google Cloud storage(Training data is uploaded to Google Cloud Storage as a CSV (comma-separated value) file). https://storage.cloud.google.com/
2. Train one of the model on the data set. https://developers.google.com/apis-explorer/#s/prediction/v1.5/prediction.trainedmodels.insert
3. Check training status of your model. https://developers.google.com/apis-explorer/#s/prediction/v1.5/prediction.trainedmodels.get
4. Apply the trained model to make the prediction. https://developers.google.com/apis-explorer/#s/prediction/v1.5/prediction.trainedmodels.predict

Note: Please use Google chrome for the Google prediction API, Generally Internet Explorer doesn’t work.
How to upload the data
1. Go to https://code.google.com/apis/console
2. Create Project ID.
3. Go to services, activate Google Cloud Storage , Prediction API
4. Go to billing on the menu at the top right of the window.
5. Swipe your credit Card. (to run the prediction based on the sample data, there is no charge, but credit card info. Is required)
6. Follow the link https://developers.google.com/prediction/docs/hello_world

Please contact me in case of problem, accessing the above link or steps.