Embedded Files
Currently graduated from AGU, Computer Engineering; and working in TurkNet. Interested with network-based deep data analytics and big data system architecture.
Education & Professional Experiences
Abdullah Gul University / Graduated Computer Engineering
2016 - 2021, Kayseri, TurkeyLectures, that I took, are Art of Computing, Object Oriented Programming, Exploring Computer Engineering, Data Structure, Mobile Programming, Digital Design, Database Management System, Software Engineering, Artificial Intelligence, Social Network Analysis, Operating System, Computer Organization, Linear Algebra, Discrete Math, Probabilities and Statistics, Korean 101 and so on.
University Of LA Verne / Student in English Language School
July 2017 - August 2017, La Verne, California, USAELS La Verne University is located in the state of California in the United States and I joined the ELS program which is one of the America's leading language schools.
University College Dublin / Visiting Researcher Intern
June 2019 - September 2019, Dublin, IrelandMainly, studied on human actions recognition through video processing by using neural network models. Also, met the requirements for summer internship programme.
TurkNet / Data Scientist Intern
July 2020 - September 2020, Istanbul, Turkey• Analyzing and creating a visualization of the Big Data given within the company and making statistics.
• Implemented machine learning algorithms to drive process optimization and improve efficiency.
• Training given Data on the practical use of Python and Machine learning algorithms within the company.
• Being able to use several Big Data visualization and editing tools, like AWS, Spark, GraphViz, Kepler.gl, Plotly.
TurkNet / Data Scientist Junior Full-time
September 2020 - August 2022, Istanbul, Turkey• Conducted extensive data analysis and visualization using SQL queries for various projects.
• Demonstrated proficiency in utilizing different analysis tools, particularly Tableau, with a focus on churn applications.
• Contributed to the backend development of web applications using Django.
• Employed ETL processes to write SQL scripts, ensuring efficient extraction, transformation, and loading of data.
• Generated ad-hoc reports tailored to specific business needs.
TurkNet / Data Scientist Mid-Level Full-time
August 2022 - August 2023, Istanbul, TurkeyExpertise in Big Data Processing Technologies and Ongoing Analysis Tasks:
• Possess extensive expertise in big data processing technologies, including handling large volume datasets (exceeding 250 GB per day on some projects).
• Successfully performed ongoing analysis tasks on various complex datasets, including complex customer and service-related CRM data, DNS data, netflow data, and geospatial data.
Building and Maintaining Analytic Reports, KPIs, Metrics, and Dashboards:
• Developed and maintained analytical reports, key performance indicators (KPIs), metrics, and interactive dashboards.
• Utilized ETL processes to write SQL scripts, ensuring efficient data extraction, transformation, and loading.
• Scripted and generated automated email reports for efficient reporting and communication.
• Leveraged BI tools such as Tableau and Microsoft Power BI for data visualization and reporting.
End-to-End Project Experience:
• Contributed to various end-to-end projects, including geographic, demographic, and behavioral customer segmentation.
• Developed an intelligent network monitoring system using ISP central POPs, enhancing network performance and reliability.
• Generated ad-hoc reports to meet specific business needs
• Supported academic projects.
Marketing Performance Analysis and ROI Optimization:
• Thoroughly examined marketing performance, user behaviors, and user conversion data funnels.
• Identified opportunities to enhance ROI, understand trends, and provided actionable recommendations for business improvement.
Self-Led with Strong Problem-Solving Skills and Emphasis on Product Development:
• Demonstrated ability to work self led, taking ownership of projects and driving them to successful completion.
Technologies: Tableau, Microsoft Power BI, MS SQL, NoSQL, MySQL, PostgreSQL, MongoDB, ElasticSearch, Python, PyTorch, PyG, NetworkX, Scikit-learn, Keras, TensorFlow, Flask, Git, Pandas, NumPy, Matplotlib, Seaborn, Spark, version control (GitFlow), CI/CD
Istanbul Technical University / PHD
September 2021 - Ongoing, Istanbul, TurkeyWill be working on:
• Data-driven Network Management for 6G systems
• Data analytics based Network Management
• Big data system architecture
Wingie Enuygun Group / Data Scientist Full-time
August 2023 - Ongoing, Istanbul, TurkeyEnd-to-End Knowledge Management Agent Development:
• Led the development of a comprehensive Knowledge Base system.
• Consolidated and organized a vast collection of internal and external documents.
• Utilized cutting-edge technologies, including AutoGen, to streamline knowledge retrieval.
• Implemented the Group Chat with Retrieval Augmented Generation for efficient information access.
• Leveraged the LangChain framework to facilitate seamless knowledge management.
• Employed the Vector DB for improved data storage and retrieval capabilities.
• Proficient in working with various AI models and APIs, including ChatGPT, Gemini, Claude, AutoGen, and MeMGPT, further enhancing knowledge management capabilities.
• One example project can be seen from enuygun.com/enbot-asistan
Analytic Report Development for Anomaly Detection:
• Designed and implemented robust anomaly detection systems for various projects.
• Utilized deep temporal neural networks to analyze different types of data and historical trends.
• Developed advanced algorithms to effectively identify anomalies.
• Improved security and performance through precise anomaly detection.
• Successfully fetched and processed data from sources like Kibana, ElasticSearch, and MySQL to support anomaly detection applications.
Optimization Expertise:
• Successfully addressed a complex shift management optimization problem as part of the Shift Management Project.
• Leveraged Google OR-Tools to optimize shift scheduling efficiently.
• Employed Reinforcement Learning-based penalizing techniques to enhance shift management outcomes.
• Developed a Flask application to streamline project operations.
• Utilized Google BigQuery to fetch and process relevant data for optimization.
Wingie Enuygun Group / AI Specialist in CTO Office
January 2024 - Ongoing, Istanbul, Turkey• Collaborating with a team of five members, including the CTO, to drive the company's technological innovation and AI First strategy.
• Organizing and managing AI projects determined every quarter, ensuring alignment with strategic goals.
• Leading the development and implementation of various AI projects, from conception to deployment.
• Coordinating with cross-functional teams to integrate AI solutions across different departments.
• Conducting regular project reviews and updates to ensure timely delivery and quality outcomes.
• Identifying and exploring new AI technologies and trends to keep the company at the forefront of innovation.
Projects
ChatBot for FAQ Page / Knowledge Base ChatBot / enuygun.com/enbot-asistan
February 2024 - Ongoing, EnuygunCreated a knowledge base chatbot with Faiss and Langchain; serviced with FastAPI.
Scoring Central POPs / Neural Network Models & Analysis
January 2023 - August 2023, TurkNetConducted near real-time analysis of customer behaviors based on ISPs central POPs scoring and published academic paper.
Customer Segmentation / Neural Network Models & Analysis
September 2023 - August 2023, TurkNetAn end-to-end project, including geographic, demographic, and behavioral customer segmentation by handling large volume datasets like exceeding 250 GB per day.
Similarities of Human Actions / Neural Network Models & Analysis / GitHub
June 2019 - May 2020, Remote, UCDSplitted videos by using (I3D) neural network model, python and its libraries like opencv. Then, found similarities between human behaviours such as standing, walking, siting etc. from video data in Google Cloud Platform and created tree by hierarchical clustering technique. Currently, working to publish an academic paper. To get more information, please contact with me.
Global Recruitment Tool / Developing Online Tool for Nokia
December 2019 - June 2020, EPIC Erasmus Project with NokiaIt's an Erasmus project, run by Nokia and aims to create an effective tools to recruit top talents from the market.
Attended “Intensive Programs for Higher Education Learners” Seminar in Hamburg, Germany.
Working remotely with an international 5 people team.
Working on a global recruitment tool project which uses Deep Learning (CNN and RNN) for scoring and parsing resumes.
Quadcopter Drone / Python & Dronekit / GitHub
November 2018 - September 2019, IEEE AGUFinished this project for competition, organised by Tubitak, with 7 people group; and played role in software part of drone project by using Linux environment, Python and its libraries like DroneKit, Numpy, Pandas etc.
Predicting Cast and Director Collaboration / Social Network Analysis / GitHub
October 2019 - January 2020, AGU Social Network Analysis LectureIt's a term project about experimental study on social network analysis and an approach that combines the graph analysis and machine learning implementations between IMDB’s cast and directors by using analysis techniques like Node2Vec, Logistic Regression, Neural Network models; data collection techniques like web scraping; and Python libraries like Pandas, NumPy.
SnapPuzzle / Android
March 2019 - May 2019, AGU Software Engineering LectureFinished an android app for mandatory course in AGU and applied Android's advanced components such as notifications, services, broadcast receivers, content providers, API, Database, Firebase etc. with 3 people group and one customer from the sector.
Non-Academic Transcript Project / Php, JavaScript / nonacademic.agu.edu.tr
2018 - 2019, AGU Youth FactoryIt's a new project that is conducted by AGU and worked with 4 people group. Created web page by using Php, JavaScript and MySQL.
Publications
Exploring the Influence of Dimensionality Reduction on Anomaly Detection Performance in Multivariate Time Series / Submitted
Mahsun Altin, Altan ÇakırThis paper presents an extensive empirical study on the integration of dimensionality reduction techniques with advanced unsupervised time series anomaly detection models, focusing on the MUTANT and Anomaly-Transformer models. The study involves a comprehensive evaluation across three different datasets: MSL, SMAP, and SWaT. Each dataset poses unique challenges, allowing for a robust assessment of the models' capabilities in varied contexts. The dimensionality reduction techniques examined include PCA, UMAP, Random Projection, and t-SNE, each offering distinct advantages in simplifying high-dimensional data. Our findings reveal that dimensionality reduction not only aids in reducing computational complexity but also significantly enhances anomaly detection performance in certain scenarios. Moreover, a remarkable reduction in training times was observed, with reductions by approximately 300\% and 650\% when dimensionality was halved and minimized to the lowest dimensions, respectively. This efficiency gain underscores the dual benefit of dimensionality reduction in both performance enhancement and operational efficiency. The MUTANT model exhibits notable adaptability, especially with UMAP reduction, while the Anomaly-Transformer demonstrates versatility across various reduction techniques. These insights provide a deeper understanding of the synergistic effects of dimensionality reduction and anomaly detection, contributing valuable perspectives to the field of time series analysis. The study underscores the importance of selecting appropriate dimensionality reduction strategies based on specific model requirements and dataset characteristics, paving the way for more efficient, accurate, and scalable solutions in anomaly detection.
Teach A Human Activity Recognition Machine Inductive Reasoning Through Hierarchical Labelling / Published
Mahsun Altin, Furkan Gursoy, Lina XuMany advanced machine learning (ML) models for video processing have been proposed for human activities recognition (HAR), along with a large number of publicly available datasets. Most existing work dedicates to improve the prediction accuracy through either creating new model structures, increasing model complexity or refining model parameters by training on larger datasets. In this paper, we have proposed an alternative idea, differing from existing work, to increase model accuracy through creating higher level summarizing labels for groups of human activities.
Intelligent Network Monitoring System Using an ISP Central Points of Presence / Published
Yousef Alkhanafseh, Mahsun Altın, Altan Çakır, Esra Karabıyık, Ekrem Yıldız, Sencan AkyüzThe proliferation of both internet usage and users have been remarkably increased due to certain situations that influenced face-to-face communications, which in turn have created high pressure on Internet Service Providers (ISPs). This research mainly aims to boost ISP services by conducting near real-time analysis for customer’s behavior movements based on their score of central Points of Presence (POP). In addition, this study focuses on establishing special Recurrent Artificial Intelligence (RNN) architecture to make daily sales predictions based on various central POPs. The process utilizes different RNN architectures, Long Short Time Memory (LSTM) and Gated Recurrent Unit (GRU), and compares them in order to make smart scoring measurements for customers’ high-dimensional data. As a result, it can be concluded that LSTM architecture has achieved much better Mean squared Error (MSE) than GRU architecture. LSTM outperforms GRU in forecasting less sensitive outliers, with an average Mean Absolute Error (MAE) of 1.354 for LSTM and 1.554 for GRU. Additionally, LSTM performs better in forecasting outliers, with an average MSE of 3.592 compared to GRU’s average of 4.8. Thereafter, the obtained results are merged over private Application Programming Interface (API) and monitored over smart reports. Eventually, the outcomes of this research can be summarized in providing several benefits for customers such as increasing internet performance, reaching promised speed, and shortening activation times. ISP-related benefits such as gaining reputation, promoting sales, and reducing customers’ negative support tickets can be achieved as well.
Volunteering Experiences
Erasmus+ / Volunteer
May 2018 - May 2018, MontenegroErasmus+, 2018 Erasmus+ activity in Montenegro, which has communication training courses. (See below)
Project Title: #OnACT
Identifier: 589976-EPP-1-2017-1-ME-EPPKA2-CBY-WB
IEEE AGU Student Branch / Board Member / www.aguieee.org
September 2018 - June 2019, AGUCompleted semester as board member in IEEE AGU student branch.
Created a portfolio website for AGU IEEE, which you can find from the below link.
Organised conferences, Arduino teaching activities, meetings etc.
Formed group of drone projects for competitions.
Google Sites
Report abuse