Konstantinos Christantonis

Konstantinos Christantonis

IT Assistant

Followers of Konstantinos Christantonis154 followers
location of Konstantinos ChristantonisGreece

Connect with Konstantinos Christantonis to Send Message

Connect

Connect with Konstantinos Christantonis to Send Message

Connect
  • Timeline

  • About me

    MSc Data Scientist

  • Education

    • Democritus University of Thrace, Greece

      2011 - 2016
      Diploma Production and Management Engineering 7.38/10

      Dissertation: Benefits and Limitations of Simulation.A literature review on the prospect of simulation tools regarding different engineering scenarios.

    • International Hellenic University

      2017 - 2019
      Master of Science - MS Data Science (8,55/10)

      Dissertation: Data Mining for Smart Cities. The thesis focuses on short-term electricity forecasting based on weather and past electricity consumption. It also offers an analysis of household appliances and the individual predictability they bring upon the total electricity load.

    • Budapest University of Technology and Economics

      2015 - 2016
      Erasmus School of Mechanical Engineering
  • Experience

    • Hellenic Army

      Sept 2016 - Jun 2017
      IT Assistant

      • IT assistant to the deputy governor of the camp. Assisted mostly on basic MS Office tasks.

    • Demetra Agriculture

      Jul 2017 - Sept 2017
      Desk Clerk

      Seasonal operator of ERP system named ‘Vegoritis’. • The main duties were to update and insert new data into the system, while visualizing the progress in sales on a weekly basis.

    • International Hellenic University

      Dec 2019 - May 2023
      Research Assistant

      Member of Intelligent Systems Laboratory and actively involved in projects regarding Natural Language Processing and Recommender Systems.• Developed a streaming framework that utilizes an online Speech-to-Text service for real time subtitles synchronization for theatrical plays. The schema includes several key practices of NLP such as character-based matching, tf-idf and LSTM. The results manage to successfully identify 95% of subtitles with mean delay less than 2 seconds.• Developed ETL components and performed evaluation on recommenders (Mean Reciprocal Rank), deployed on retail e-shops, while also engaged on customers shopping behavior prediction (purchase intent) through sequential models such as LSTM and GRU with binary accuracy higher than 97%.• Designed and developed a recommendation framework that crawls data and applies content and graph-based methods such as node2vec for the needs of a tourism agency chatbot.• Developed an automatic information extraction tool (structure/responsibilities) on Greek Government Gazette documents, through deep learning (NER,Question-Answering) and rule-based approaches. The ML models outperform the rule-based implementations, achieving 91% accuracy on the structure task and 61% on the responsibilities task. Show less

    • Workearly Greece

      Sept 2023 - Aug 2024
      IT Consultant

      • Contributing actively to the R&D team, with a focus on developing applications using Large Language Models for sports analytics and educational purposes.• Developing new learning materials for Data Science and Machine Learning courses, concurrently maintaining existing content for programming languages, SQL/Python.• Conducting live sessions to ensure clients' understanding and proficiency in the learning process.

  • Licenses & Certifications