Kemal Gunay
Postdoc Computational Social Scientist
University of Trento
Biography
I serve as a Postdoc Data Scientist at Trento University’s Sociology Department. My research interests include user experience, public opinion mining, discourse analysis, social psychology, and new media. As a data nerd, I spend most of my time analyzing large-scale social media data to better understand how information and behavior spread through society. My expertise lies in social media analytics, where I study NLP, complex network structures, such as relationships between people in social networks, or politicians and their audiences.
I obtained my PhD - thesis and master’s degree from Istanbul University as well as master’s degree in corporate communication. I’ve held a post-doctoral scholar in Trento University’s Sociology Department.
I donate at least 10% of my income to effective charities. Perhaps I could inspire you to do the same!
Download my academic CV and resumé .
Interests
- Computational Social Science
- Climate Change Communications
- Quantitative Text Analysis / NLP
- Statistics
- Media Studies
Education
-
PhD in Communication Sciences, 2022
Istanbul University
-
Communication and Social Sciences (Erasmus)
Universidad San Jorge (Spain)
-
Corporate Communication
Istanbul University
-
Music Technology
Bournemouth and Poole College (United Kingdom)
-
Public Relations and Publicity
Istanbul University
Skills
R
I have 4 years of experience using R for academic works
Statistics
Python
I have 4 years of experience using Python for data science projects
JavaScript
I have begun learning JavaScript this year, specifically using it to scrape data from social media
Docker
AWS
Experience
Postdoc Data Science Researcher
Responsibilities include:
- Field of Study: Research Methods in Digital Media, Data Science, Social Media Analytics
- Analysis Methods:: Collecting and managing large databases from major social media platforms; using quantitative analysis methods and techniques; specifically (but not exclusively) analyzing data using techniques from computational linguistics (NLP), network science (social and/or semantic), and machine learning.
- Academic research tools: Python & R, data manipulation, data visualization, statistical tests.
Lecturer
Responsibilities include:
- Field of Study: Research Methods in New Media, Data Science, Social Media Advertising
- Analysis Methods:: I teach core courses in the social sciences: research methods and statistics
- Academic research tools: Python, RStudio, SPSS, NVivo Software
Ph.D New Media Researcher (Scholar)
Responsibilities include:
- Field of Study: Digital Media, Environmental Communication and Communication Sciences, Data Science, Political Discourse
- Analysis Methods: Text mining & NLP; Topic Modelling (LDA, STM), Social Network Analysis, Text Clustering
- Academic research tools: Python, RStudio, SPSS, NVivo Software
Data Scientist
Responsibilities include:
- Performing data extraction, manipulation, feature engineering and visualization. Applying statistical testing and ML techniques: hypothesis testing, classification, regression.
- Working with ML algorithms and models (SVM, NB, RF, LR, PCA, Rule- Based) and their underlying computational and probabilistic statistics.
- Hands-on-experience via projects; CRM Analytics; AB Test, Recommendation Systems; Measurement, Regression, Classification and Time Series Problems, Demand Forecasting and et al.
Researcher
Responsibilities include:
- Field of Study: Digital Media and Communication Sciences, NLP & Text Mining, Data Visualization, Machine Learning
- Academic research through Python, RStudio, SPSS, NVivo Software
Projects
*
Association rule learning is a rule-based machine learning approach for finding significant connections between variables in large databases. It is designed to identify strong rules that have been identified in databases using various measures of interests tool that can help businesses and other organizations manage their interactions with customers.
Constructing social networks from unstructured text is an issue of great interest that receives little research. In my most recent blog article, I describe how I built a social network.
Lifetime value is a critical metric because it represents the maximum amount that customers may be expected to spend in order to acquire new ones. As a result, it’s crucial in determining the payback of marketing expenses used in marketing mix modeling.
In this notebook, We will discuss three types of recommender system; (1)Association rule learning (ARL), (2)content-based and (3)collaborative filtering approaches. In this notebook, we will explain how to build a recommender system with these three methods.
Building knowledge graphs from unstructured text is an issue of great interest that receives little research. In my most recent blog article, I describe how I built a knowledge network in Neo4j using the LangChain framework and OpenAI functions. Along with learning all the subtleties of employing LLMs to build a knowledge graph, you will also learn the necessary procedures to develop a production information extraction pipeline.
Customer Relationship Management (CRM) system is an information management and analysis tool that can help businesses and other organizations manage their interactions with customers.
Deep learning (also known as deep structured learning) is part of a broader family of machine learning methods based on artificial neural networks with representation learning. Learning can be supervised, semi-supervised or unsupervised
In this task, you are required to create an algorithm that takes as input a pdf file corresponding to a research publication and outputs a list of all geographical locations mentioned in the publication. For each geographical location, the algorithm will have to additionally identify the country that the location belongs to, and return a latitude- longitude pair corresponding to the centroid of the respective country.
A few problems associated with image data include complexity, inaccuracy, and inadequacy. This is why before building a computer vision model, it is essential that the data is preprocessed (cleaned and processed to the desired format) to achieve the desired results.
Republic Day (29 October 1923) Republic Day (Turkish Cumhuriyet Bayramı) is a public holiday in Turkey commemorating the proclamation of the Republic of Turkey, on 29 October 1923. The annual celebrations start at 1:00 pm on 28 October and continue for 35 hours.
Family of linear algebra algorithms for identifying the latent structure in data representAs we have previously shown, machine learning may be used in a variety of ways to automatically classify texts or documents.
Customer Relationship Management (CRM) system is an information management and analysis tool that can help businesses and other organizations manage their interactions with customers.
Family of linear algebra algorithms for identifying the latent structure in data represented as a non-negative matrix. NMF can be applied for topic modeling, where the input is term-document matrix, typically TF-IDF normalized. Input; Term-Document matrix, number of topics. Output; Two non-negative matrices of the original n words by k topics and those same k topics by the m original documents.Basically, we are going to use linear algebra for topic modeling.
How often do you use the RFM analysis in your marketing? The RFM analysis is a great tool for marketers to figure out which customers are most valuable and how they should be marketed to. It’s also a great way to find new prospects that might not have been found otherwise. Here we’ll explain into each of the four groups. Retention, Frequency, Monetary Value and Referral and explain what they mean and why you want them as customers!
Companies want their data science teams to speed up the process so they can deliver valuable business predictions faster. That’s where ML pipelines come in. By automating workflows with machine learning pipeline monitoring, ML pipelines bring you to operationalizing machine learning models sooner.
Social network analysis (SNA) is the process of investigating social structures through the use of networks and graph theory. It characterizes networked structures in terms of nodes (individual actors, people, or things within the network) and the ties, edges, or links (relationships or interactions) that connect them.
Spotifyr is an R wrapper for pulling track audio features and other information from Spotify’s Web API in bulk. You can easily analyze your listining records.
Republic Day (29 October 1923) Republic Day (Turkish Cumhuriyet Bayramı) is a public holiday in Turkey commemorating the proclamation of the Republic of Turkey, on 29 October 1923. The annual celebrations start at 1:00 pm on 28 October and continue for 35 hours.
Recent Publications
Quickly discover relevant content by filtering publications.
(2022).
Topıc Modelıng Analysıs Of NGO’s Twıtter Postıngs Between 2020-2021 In Turkey Wıthın The Context Of Clımate Change Communıcatıon.
The Turkish Online Journal of Design Art and Communication, 12(4).
(2021).
Digital Seige - Is the Internet of Things Transforming a Surveillance Tool?.
Istanbul University Press, 5281(5281).
(2021).
Value-based Communication During COVID-19 Pandemic - A Study on The Twitter Messages of Turkish Ministry of Health.
Athens Journal of Mass Media and Communications, 7(1).
(2020).
An Investigation of Candidate Leaders’ Tweet Campaigns Prior to the Istanbul Metropolitan Municipal Elections Using Big Data Text Mining.
Connectist: Istanbul University Journal of Communication Sciences, 59(59).