Noblis’ Post

Exciting News! 🎉 Our research paper titled "Efficient Task Allocation with Mentorship Mapping" has been accepted for presentation at the prestigious 7th International Conference on Computational Intelligence in Data Science (ICCIDS 2024)! We're thrilled to share our insights on optimizing task allocation and leveraging mentorship mapping to enhance productivity and knowledge sharing. Join us at ICCIDS 2024 as we delve into the latest advancements in computational intelligence and data science. Stay tuned for updates as we prepare to showcase our work at this esteemed conference! #ICCIDS2024 #ResearchPaper #TaskAllocation #MentorshipMapping #DataScience #ComputationalIntelligence

#Flashback to the #DARESymposium keynote delivered by Professor Scott Sisson, a globally recognised expert in computational and Bayesian statistics. Scott's enlightening presentation showcases some of the work by his recent PhD students in “symbolic data analysis”, which aims to perform statistical analyses for data that have been summarised into distributional form, such as random rectangles, random histograms, and other distributions. Be sure to watch the video for further insights! UNSW Data Science Hub (uDASH) | #DataScience #DataAnalysis #Statistics #FlashbackFriday

Thrilled to have participated in Data Day, hosted by the Faculty of Science at Carleton University for Data Science (CUIDS). This annual conference is a testament to the evolving field of data science and analytics, showcasing cutting-edge research and industry insights. Our team had the opportunity to present a poster delving into an in-depth analysis of Ottawa's road accidents from 2017 to 2023. Our study, which aligns with the United Nations' road safety goals, examined various factors influencing accident trends and identified critical areas for intervention to enhance road safety in Ottawa. Data Day 10.0 was an enriching experience, featuring presentations from Sprott School of Business at Carleton University Prof. Mohamed Al-Guindy and Carleton's esteemed researchers and industry professionals, a keynote address, and a panel discussion that fostered meaningful conversations on the future of data science. Special thanks to Majid Komeili from the School of Computer Science for his insightful contributions during the panel discussions, focusing on the next decade of data science, with a spotlight on explainable AI, healthcare, and environmental sustainability. I'm also grateful to Professor Elio Velazquez and my dedicated team members - Ian Hlomador, Hieu Le, and Zahra Mousavi - for their collaboration and hard work. The poster competition was a highlight, allowing us to engage with peers and experts, sharing our research and gaining valuable feedback. Looking forward to more such inspiring events that bridge academia and industry, pushing the boundaries of data science to new horizons. #DataDay10 #DataScience #CarletonUniversity #RoadSafety #ResearchImpact

What is your opinion on the academic #digitaltransformation? 👾 Join our HIDA lecture series with Dr. Melissa Laufer on February 14 and discuss with us the impact of Large Language Models on Academics and Academic Work!

🔍 Excited to share my latest blog post on t-SNE! 🌟 In my recent exploration of dimensionality reduction techniques, I've delved into the fascinating world of t-SNE (t-distributed stochastic neighbor embedding). 🚀 In this blog, I demystify t-SNE and its powerful ability to visualize high-dimensional data in lower dimensions while preserving local relationships. From its graceful handling of distance to the control of density with the variance parameter, I've covered it all! 💡 If you're curious about unlocking complex patterns hidden within your data or want to enhance your data visualization skills, this blog is a must-read. Dive in and discover how t-SNE can revolutionize your understanding of data! 📊✨ Read the full blog here https://lnkd.in/g4AyHuvH. #DataScience #MachineLearning #DataVisualization #tSNE #DimensionalityReduction #DataAnalysis

Exploring t-SNE vs. SNE in Simple Terms When it comes to visualizing complex data, t-SNE and SNE are two popular methods. Let's break down the differences: 🔹 SNE (Stochastic Neighbor Embedding): ▪ Keeps Neighbors Close: SNE focuses on keeping similar data points close together. ▪ Slower for Big Data: Takes more time and power to process large datasets. ▪ Crowding Issue: Can struggle to show data clearly when reduced to 2D or 3D. 🔹 t-SNE (t-Distributed Stochastic Neighbor Embedding): ▪ Better Visuals: Uses a special method to prevent the crowding problem, making clusters more distinct. ▪ Faster with Large Data: More efficient at handling bigger datasets. ▪ Widely Used: Preferred for creating clear and insightful visualizations of data. Why Choose t-SNE? 🔹 Speed: t-SNE is quicker for large datasets. 🔹 Clarity: Produces clearer and more meaningful visuals. 🔹 Ease of Use: Simple to implement and understand. #DataScience #MachineLearning #DataVisualization #tSNE #SNE #DataAnalysis

On Day 3 of 30 Days of Statistical Modelling, lets talk about a very interesting measure of dispersion. The inter quartile range. The interquartile range (IQR) is a robust statistical measure that provides insights into the spread of data by focusing on the middle 50% of a dataset. Calculated as the difference between the third quartile (Q3) and the first quartile (Q1), the IQR is less sensitive to extreme values, making it particularly useful for assessing variability in skewed distributions. Understanding the IQR enhances our ability to grasp the central tendencies and overall distribution of data, crucial for making informed decisions in various fields. #statsitics #datascience #communitylearning

Navigating the Data Preprocessing Journey! 🛠️ My dissertation's comprehensive methodologies prepare data for effective analysis and modeling. 📊 Each step—feature selection, categorical feature encoding, scaling, and resolving class imbalance—plays a significant role in refining my dataset. Chi-square tests helped me identify less useful factors. 🌐 Witness the transformation of raw data into a well-prepared dataset, establishing the groundwork for powerful prediction models! #DataPreprocessing #MachineLearning #DataScience #LinkedInAnalyticsInsights.

🔍 Unlocking the Hidden Patterns in Complex Data 📊 with t-SNE! Discovering insights from high-dimensional data can be a challenging task, but fear not! 🚀 t-SNE (t-distributed Stochastic Neighbor Embedding) is here to the rescue! 🌟 This powerful unsupervised non-linear dimensionality reduction technique enables us to explore and visualise complex datasets, transcending the limitations of straight-line separations. 💡 By condensing data into two or three dimensions, t-SNE grants us an intuitive understanding of how information is arranged in higher realms. Are you ready to unravel the secrets hidden within your data? 🤔 With t-SNE, you can gain valuable insights into underlying patterns and relationships, making data exploration a breeze! 🌈 Image src: DataCamp. #DataScience #tSNE #DataExploration #DimensionalityReduction #Visualisation #DataInsights

Out in EPJ Data Science, I discuss: "What can the two sides of #ComputationalSocialScience, i.e., social science and computational science learn from each other in thinking spatially?" Motivated by Yong-Yeol Ahn's brilliant keynote IC2S2 2021 Open Access here: https://lnkd.in/em6VqJUs Many thanks to Xinwei Xu (ETH Zurich), Ivan Smirnov, Erin Walk and Andres Castro for insightful discussions, careful reading and comments on the draft, and Termeh Shafie and Christoph Stadtfeld for organizing the special issue.