

Data Skeptic
Kyle Polich
The Data Skeptic Podcast features interviews and discussion of topics related to data science, statistics, machine learning, artificial intelligence and the like, all from the perspective of applying critical thinking and the scientific method to evaluate the veracity of claims and efficacy of approaches.
Episodes
Mentioned books

Oct 14, 2024 • 30min
Network Analysis in Practice
Asaf Shapira, a network analysis consultant and the host of NETfrix, dives into the intricacies of network science. He discusses how network analysis techniques can identify malicious activities, including bot farms on social media. The conversation touches on the impact of social networks on elections and the importance of community detection algorithms in understanding organizational dynamics. From the historical roots of network analysis to its modern applications in areas like COVID-19 contact tracing, Asaf sheds light on its widespread yet underutilized potential.

Oct 7, 2024 • 30min
Animal Intelligence Final Exam
This discussion wraps up the Animal Intelligence season with reflections on production challenges and guest engagements. The speakers emphasize how seamless editing enhances listener experience, especially for non-native English speakers. They delve into the intersection of machine learning with biology and the significance of integrating these technologies for species identification. The ethics of animal research and the complexities of animal consciousness are pondered, alongside personal experiences that highlight ongoing education in data science.

10 snips
Sep 24, 2024 • 26min
Process Mining with LLMs
David Obembe, a recent graduate from the University of Tartu, dives into his master's thesis on blending large language models with process mining tools. He explains how process mining uses event logs to map out inefficiencies in business processes. Fascinating insights include the evolution of these techniques post-LLM integration, enhancing data retrieval and insights. David shares his experiments with Retrieval Augmented Generation and discusses the challenges of prompt engineering, highlighting the balance between accuracy and model reliability.

Sep 17, 2024 • 23min
Open Animal Tracks
Risa Shinoda, a PhD student from Kyoto University focusing on computer vision, dives into the fascinating world of animal tracking. She unveils the OpenAnimalTracks dataset, designed for predicting animal footprints and discusses her model’s algorithms and accuracy. Risa explores how computer vision is revolutionizing agriculture, enhancing practices and animal welfare. She also addresses challenges in capturing precise photographic evidence and the critical role of understanding animal tracks in wildlife conservation.

12 snips
Sep 10, 2024 • 40min
Bird Distribution Modeling with Satbird
Mélisande Teng, a PhD candidate at Université de Montréal, dives into her groundbreaking research on biodiversity monitoring using remote sensing and computer vision. She discusses the innovative Satbird project, which enhances bird distribution modeling by combining satellite data and citizen science. The conversation highlights challenges like data imbalance in different regions and the importance of acoustic monitoring. Mélisande also explores the intricacies of joint species distribution modeling and advocates for collaboration between machine learning and ecology to advance conservation efforts.

Aug 26, 2024 • 31min
Ant Encounters
In this discussion, Deborah Gordon, an author exploring ant colony dynamics, shares fascinating insights on how these tiny creatures exhibit complex behaviors. She highlights the concept of collective intelligence, revealing how simple interactions lead to adaptive solutions. Deborah also discusses the vast diversity of ant species and their unexpected habitats. Listeners will learn how ants' navigation strategies and social interactions can inform models in artificial intelligence, showcasing the remarkable intelligence of decentralized systems.

7 snips
Aug 19, 2024 • 39min
Computing Toolbox
Madlen Wilmes, co-author of 'Computing Skills for Biologists,' shares her insights on essential computing skills for biologists. She discusses her transition from academia to finance, highlighting how transferable data analysis skills can open unexpected career paths. Madlen emphasizes the importance of programming languages like R and Python in data science. She also covers the challenges and advantages of moving into industry, including the need for strong networking and collaboration skills, while addressing the impact of soft skills on professional success.

Aug 14, 2024 • 32min
Biodiversity Monitoring
Hager Radi, a specialist in biodiversity monitoring, delves into the intricate world of species distribution modeling. She discusses the challenges posed by incomplete data and biases in presence-only datasets. Hager highlights the innovative use of machine learning and remote sensing, showcasing how these technologies can help predict species distributions even with limited observations. She also sheds light on exciting developments like using drones and citizen science platforms, emphasizing the importance of tech in conservation efforts.

Aug 8, 2024 • 41min
Hacking the Colony
Ashay Aswale, a swarm robotics researcher inspired by insects, and Tony Lopez, who highlights challenges faced by robotic swarms, delve into the fascinating intersection of nature and technology. They explore how robotic swarms can learn from ant colonies, tackling issues like misinformation and communication breakdowns. The duo discusses the advantages of specialized roles within swarms and the real-world applications in areas such as space exploration. Their insights reveal how understanding ant behavior can enhance the resilience and efficiency of robotic systems.

Jul 31, 2024 • 33min
Primate Poses
In this insightful discussion, PhD student Richard Vogg shares his pioneering work on tracking lemurs and macaques using multi-camera setups. His research focuses on automating behavioral analyses in the wild, revolutionizing how we understand primate behavior. Vogg elaborates on the challenges of maintaining accurate tracking with uncalibrated cameras and the advantages of using advanced computer vision technologies. Listeners will discover how these innovations enhance scene understanding and allow for more reliable identification of individual animals.


