Freelancerı Projeye Davet Edin
Görünüşe göre şu anda aktif bir projeniz yok. Neden şimdi bir proje ilan etmiyorsunuz? Ücretsizdir!Bir Proje İlan Edin
- N/ATamamlanmış İşler
- N/ABütçe Dahilinde
- N/ATekrar İşe Alım Oranı
Python scripts for Time Series - Machine Learning with TensorFlow
“did not deliver”Creatory 1 yıl önce
Software Development InternJan 2017 - Mar 2017 (2 months)
Built a Deep Neural Network(Convolutional LSTM) to analyse the flow of emotion in the text being typed by the user. The model was built using GPU-enabled TensorFlow. Served the model as a Web Service and used it in the Windows Software Keyboard. The served model is available for public use on [login to view URL]
Consultant Data ScientistSep 2016 - Nov 2016 (2 months)
Worked on building a self-help intelligent product for banks using Machine Learning and Web Services. This allowed the banks to build their own predictive analytics models, by just interacting with the interface and not with the intricate math involved in Machine Learning.
Data Science InternJun 2016 - Aug 2016 (2 months)
Built an Intelligent Credit System, using Pandas, Scikit-Learn, Numpy and Pickle, used various Machine Learning Classification Techniques for building the model and then used a Hard Voting Technique to further increase the Accuracy. The Built Model was then provided for real-time usage, by using Web Services, dished out using a Django WebServer and JSON used as Data Interchange format.
Summer Research InternJun 2015 - Jul 2015 (1 month)
Devised a novel text compression algorithm using frequent pattern mining and hash tables, presented our work at The First EAI International Conference on Computer Science and Engineering, Penang Malaysia. The proceedings are published by European Union Digital Library.
Bachelor of Technology, Information Technology2013 - 2017 (4 years)
Triah: an intelligent guiding system for the visually impaired
To improve the quality of the lifestyle that the visually impaired possess, we propose an assistive model which combines the various aspects of computer vision. Our proposed model aims at detecting the number of faces by using Haar cascade classifiers and integrating it with Raspberry Pi. The processed images are run through the classifier, and the user is notified of the spatial orientation of the people surrounding them via headphones. DOI:10.1007/s40012-016-0111-2
Hash Based Frequent Pattern Mining Approach to Text Compression
The paper explores the compression perspective of Data Mining. Huﬀman Encoding is enhanced through Frequent Pattern Mining. The seminal Apriori algorithm has been modiﬁed in such a way that optimal-number of patterns(sequence of characters) are obtained. These patterns are employed in the Encoding process of our algorithm, instead of single character based code assignment approach of Conventional Huﬀman Encoding. DOI:10.4108/eai.27-2-2017.152268