Veri Bilimi DEV - Videos
In this video, I leverage advanced regression techniques to predict house prices in Boston. I believe this video is particularly useful for those interested in participating in Kaggle competitions. As mentioned in my previous videos, I created this content to share my experiences in data science, especially addressing the challenges I faced due to a lack of Turkish resources and issues with working code. I hope this video proves to be educational and insightful.
In this tutorial, I'll explain how to apply artificial neural networks to e-commerce data, with a focus on practical implementation. Additionally, this video serves as a guide on how to upload a 'notebook' to Kaggle, providing a step-by-step example. The final goal is to demonstrate how to generate all the coding aspects using ChatGPT4.0. I'm also sharing this knowledge considering the scarcity of Turkish resources in Data Science.
In the Kaggle notebook, I'll address Amazon UK product statistics. The dataset contains some empty and n/a cells. I'll perform exploratory data analysis to derive insights, followed by applying an artificial neural network. The process includes cleaning the dataset by removing duplicates in the ASIN and BSR columns, and empty cells in the Sales column. The target variable is Sales, with independent variables including Price, Revenue, Review Count, Ratings, BSR, Image, Weight, Active Sellers #, and Review Velocity. The ANN model will use two layers with a Feed Forward approach, nine neurons for independent variables, and a sigmoid function as the activation function.
In this video, I delve into the "car prices" keywords searched on Google Search Trends, employing Time Series Analysis to forecast the search frequency of these terms. I utilized the Autoregressive Integrated Moving Average (ARIMA) method for this estimation. As I've mentioned in previous videos, my journey in learning Data Science was challenging due to the lack of Turkish resources and frequent issues with working code. This video is my attempt to share the experiences and knowledge I've gained in this field. I hope it serves as an informative and educational resource.
In this video, I explore the processes of linear regression using the Germany Housing dataset. Similar to my previous videos, I created this content to share my experiences in the field of Data Science, particularly emphasizing the challenges I faced due to a lack of Turkish resources and issues with working code. This video is aimed at providing an educational perspective on linear regression techniques.
James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An Introduction to Statistical Learning (Vol. 112, p. 18). New York: Springer.
I hope you find this video informative and useful.
In this video, I analyze the Central Bank Committee Decision Summaries using the Sentiment Analysis method. The choice of the Central Bank of Turkey (T.C Merkez Bankası) stems from the realization that we often don't observe the impact of institutions we interact with in our daily lives. Specifically, I explored how the decisions of a country's central bank affect our lives, a question I hadn't previously contemplated.
In the video, I delve into the emotional connotations of words in the bank's decision summaries using Sentiment Analysis. As in previous videos, I created this content to share my experiences in learning Data Science, especially considering the lack of Turkish resources and challenges with working code.
I hope this video is educational and insightful.
References for further reading and tools:
In this video, I focus on the Principal Component Analysis (PCA) method, a technique in unsupervised learning, using a dataset of the 50 most listened-to Spotify songs in 2019. I created this video to share my experiences in data science, particularly addressing the challenges I encountered due to a lack of Turkish resources and difficulties with working code. This video is designed to be educational, providing insights into the application of PCA in analyzing real-world data sets.
References for further reading and tools:
In this video, I introduce 'Veri Bilimi DEV', a GPT model I created using the "Create a GPT" feature, incorporating my learnings and experiences as a Data Scientist. Veri Bilimi DEV is a program primarily designed to analyze Excel files, especially focusing on data from Amazon marketplaces. It features Amazon Market Research, Amazon Product Review Analysis, Amazon PPC reports analysis, and Amazon Machine Learning reports using PCA (Principal Component Analysis) and Classification Tree models.
Despite its limitations as an independent program from ChatGPT, Veri Bilimi DEV is a productive tool in data science. I've also instructed it to self-guard against cyberbullying. As always, this video is born out of my journey in learning Data Science amidst a scarcity of Turkish resources and challenges with working code. I hope you find this video educational.