Data Scientist İstanbul, Türkiye

Kan vara tillgänglig

(Uppdaterat 2021-12-27)

Data Scientist

İstanbul, Türkiye

Modersmål English

  • + 5 years of experience in Deep Learning
  • + 5 years of experience in Python
  • + 5 years of experience in Pandas

Kompetenser (11)

KERAS

Flask

Python

Data Analysis

DATA SCIENCE

Pytorch

GCP

TENSORFLOW

Docker

NEURAL NETWORK

Spark

Sammanfattning

Mainly Programming on Python (5 Years) used the libraries such as scikit-learn, pandas, keras, Flask, NLTK, spaCy. Currently, improving on tensorflow.
Other Languages: C# (2 Year). Java (3 Years), MATLAB (1 Year) 
 

Professionell bakgrund

Data Scientist
Dcipher Analytics

2021-02 - Pågående

Co-operated with UC Berkeley students and guided them for a project where we achieved to create a solution to generate a knowledge graph from unstructured text.
• Collaborated on the implementation of SurveyMonkey API to the platform.
• Fine-tuned natural language models such as GPT2 for text generation.
• Currently working on the backend team to implement new NLP solutions on the platform.
Data Scientist
Cerebro

2019-08 - 2021-02

With the Software Development team, I acted as a gateway between the Data Science and Software Development team, helped with the maintenance and development of the product based on the needs of the Data Science team.
• With the Data Science team, I analyzed the behaviours of hotels/regions for several OTAs and created regression models with my teammates.
o I created jupyter notebooks for the mentioned OTAs, analyzed and visualized the data using the libraries such as pandas, matplotlib, seaborn, plotly.
• Created the virtual environment(s) on the virtual private servers and helped with the maintenance.
• Currently, working on a game-based project.
o Analyzing user behaviour and creating models for churn prediction, customer lifetime value and so on.
Machine Learning Engineer
Talentra Recruitment Agency

2018-10 - 2019-06

Studied about Natural Language Processing (NLP) and its applications on Python with the libraries NLTK and spaCy.
• Used Neo4J as the database to hold the information and build the relationship between entities.
• Used tf-idf vectorisation and SVM Classifier to predict the expertise of people from a given CV. This study is done with Scikit-learn library.
• Finished the project which has been funded by TUBITAK "Profiling and Talent Extraction System with Semantic Matching".
• Heroku is used as the Cloud Platform and Flask is used to build the APIs for the platform.
• Created a Convolutional Neural Network (CNN) model to determine a given CV's format with 90% accuracy using Keras.

Akademisk bakgrund

MSc - Computer Science
University of Birmingham

2017-09 - 2018-09

BSc in Electrical and Electronics Engineering
Istanbul University

2013-09 - 2017-07

Certifikat

Deep Learning Specialization by Coursera

Kontakta konsult

/