I build some awesome web crawler (spider, bot) which can scrape the data from the web and store it into the database and can be used for some analysis/machine learning work or some business intelligence purposes. I can automate the web crawling task which can be scheduled will scrape the particular data from time to time.
If you already have some useful data in the database, It can be useful for business intelligence purposes. I can help you get some useful insight from the data using data mining techniques and machine learning algorithms, apart from that I will also suggest you what are the other things you can collect to improve the business which will help you to analyse the trends and insight more effectively. I have used Machine learning algorithms for different type of classification (multiclass classification, multi-label classification, hierarchical classification) tasks in the real scenario, built a text analyser and much more.
I have experience in many AWS Services like (EC2, Lambda, S3, SNS, SES) and have used the services in many of the platforms.
I know how to build API using Flask framework of python, and Also, I have build some to deploy the Machine Learning Models.
I like to play with data, analyse the data, built solution using latest technology which are very efficient and reliable. With improvement in technology and new research in the field I have to stay updated with the latest trends in the technology. So, I keep learning a lot about new technology and keep myself in loop and leverage the improvement in the technology for the benefits of the company as well as for self-growth.
Currently working developing the in-house solutions for various purpose and build and IT infrastructure for the Sameeksha Capital from scratch.
July 2022 to Current
October 2020 to March 2022
July 2019 to October 2020
January 2019 to July 2019
I write here, How to do this/that things and some of the useful content which can be useful as a guide (Basics to start with) to the readers out there to start with python and build some awesome solutions. Check out Egrasps Publication on Medium. Also I write as contributor to the Data Driven Investors (DDI) Main website. Checkout My posts on DDI Website.
Sentimental analysis of Twitter Data for a particular product/keyword. You can fetch the tweets related to keywords and also after analysis it will tag the tweet with negative, positive and neutral. So, which can be filtered and we can drive some change to our method or product according to suggestion and negative comments. Which help to grow business by making the positive changes to our product.
In this project, I used Scikit-learn to perform a decision tree based classification of weather data. Weather data is used is from location San Diego, California. Data was collected for a period of three years, from September 2011 to September 2014, to ensure that sufficient data for different seasons and weather conditions are captured. I have performed the use of the Decision tree classifier of the Scikit-learn library of the python and using the Decision Tree Classifier it will predict if there was a high humidity level or not based on some other weather factors.
In this project I have demonstrated the web scraping task from the website Craigslist and find some event list which are going to take place in the city mumbai. And then save the output to the file in the form of ,csv file. Find step by step guide to achieve this task.
In this project, I have demonstrated the Natural Language Processing Of Movie Reviews using NLTK. Do the semantic analysis of the movie reviews out of 2000 reviews which are the words that decide that the movie is having a positive review or negative one. I used the Nltk library of python for the same. I used a Naive Bayes classification algorithm of the Nltk library with train data. Then test its accuracy using remaining test data. And resultant accuracy of the model is around 70 percentage.
In this project, I have demonstrated the web scraping using request, scraped the event data from some of the event listing platforms and then parsed the scraped data using Beautifulsoup. Then organise it into Dataframe and perform the text similarity analysis Gensim and Gensim's TF-IDF similarity checker based on TF-IDF (term frequency-inverse document frequency) and list out the unique titles of the events.