![Duke Center for Computational Thinking](/img/default-banner.jpg)
- Видео 141
- Просмотров 103 193
Duke Center for Computational Thinking
США
Добавлен 28 окт 2021
Identity and Computation: Using Art to Question the Limitations of Machine Learning
Duke University Ph.D. student in Computational Media, Arts & Cultures Hugo Idarraga Franco discusses his thesis, which challenges the idea of some machine learning practitioners that the computability of the world is doable. His hypothesis is that machine learning indeed has limitations for classifying the world, which he tests through an interactive art installation.
Art installation footage courtesy of Yasaman Baghban.
Art installation footage courtesy of Yasaman Baghban.
Просмотров: 72
Видео
Environment & Health: Analyzing Air Pollution Effects During Pregnancy on Newborn Gene Expression
Просмотров 632 месяца назад
Duke Ph.D. candidate in the Nicholas School of the Environment and Computational Fellow Reshma Nargund discusses her project, which studies the effects of air pollution during a mother's pregnancy on the expression of newborn genes.
Analyzing Ocean Maps: Using New Data Technologies to Understand Ecological and Governance Challenges
Просмотров 602 месяца назад
Duke Ph.D. candidate in the Nicholas School of the Environment's Marine Science & Conservation program and Computational Fellow Emily Melvin discusses her project analyzing ocean maps and using data technologies to explore and understand the ecological and governance challenges they bring to light.
Training AI: Fine-Tuning a Chatbot for Duke
Просмотров 743 месяца назад
Duke University and Code program participant Jonathan Reyes discusses how he and his team trained a chatbot using Duke-specific information.
EcoBin: Using Image Recognition to Sort Recyclables
Просмотров 763 месяца назад
Duke University student Mick Tobin (class of 2025) describes the challenges and triumphs of using image recognition software to sort waste.
Solving with Software: Streamlining Package Pickup for Students
Просмотров 1675 месяцев назад
Duke University Code program participant Alex Pieroni (class of 2025) describes how he and his team developed a tool to address the issue of waiting in long lines to pick up packages at the Campus Mail Center. Begun as a Code (codeplus.duke.edu) project, the product is being piloted at Duke with the intention of moving it to production.
Revolutionizing Data Analysis with AI and AWS
Просмотров 1066 месяцев назад
Duke Master's in Electrical & Computer Engineering student Isaac Wang describes how his summer 2023 internship project transforms data analysis for scientific research making it more accessible, precise, and economical using AI and Amazon Web Services (AWS).
Neurocities and Ruinscapes: The Neuroscience of Eye Movement and Archeological Site Exploration
Просмотров 1318 месяцев назад
Eric Rios Soderman - a masters student in Interdisciplinary Studies and 2023 Data summer program participant discusses his team's project to track what people are looking at when they look at an archeological ruin.
Auditory Imagery: Using EEG to Distinguish Among Sound Categories
Просмотров 3208 месяцев назад
Julia Leeman a neuroscience and music double major (Duke class of 2024) and 2023 Data summer program participant discusses how her team employed machine learning of EEG data to study how the brain distinguishes among categories of sounds.
Taming the Beast: The Need for a Thoughtful AI Governance
Просмотров 1029 месяцев назад
Dr. Michael Pencina, Chief Data Scientist for Duke Health and Director of Duke AI Health, presents Taming the Beast: The Need for a Thoughtful Health AI Governance. Dr. Pencina shares strategies, pitfalls, and opportunities for healthcare data science, including the importance of improving equity. Recognized as an international authority on risk algorithms and their evaluation and governance, D...
Lifesaving Potential: Using Machine Learning to Screen for Breast Cancer
Просмотров 10410 месяцев назад
In this video, Duke Masters in Medical Physics student Kyle Ferguson discusses the potential of machine learning to screen for breast cancer. Kyle was a 2022 AI Health Machine Learning Summer School participant (aihealth.duke.edu/2022/04/18/duke-machine-learning-summer-school-2022/).
Exploring ChatGPT: Real World, Less-Controversial Uses for Everyday Life
Просмотров 8410 месяцев назад
Stephen Toback, Assistant Director of Academic Media Production and Engineering in Duke's Office of Information Technology, presents a compelling presentation on the advancements in generative artificial intelligence (AI), specifically focusing on chatbots. If you're unfamiliar with AI chatbots or are experienced and want to learn about new ways to leverage this technology, this session is wort...
Applying For Credit In A Digital World
Просмотров 3811 месяцев назад
Applying For Credit In A Digital World
Demystifying Cloud Computing: Creating Resources for Duke Users
Просмотров 102Год назад
Demystifying Cloud Computing: Creating Resources for Duke Users
Early Modern London Sermons: Using NLP to Determine Attitudes About Charitable Giving
Просмотров 80Год назад
Early Modern London Sermons: Using NLP to Determine Attitudes About Charitable Giving
2023 Health Data Science Poster Showcase
Просмотров 99Год назад
2023 Health Data Science Poster Showcase
Design with the Other: What Do We Mean by Equity Centered Design?
Просмотров 213Год назад
Design with the Other: What Do We Mean by Equity Centered Design?
Working with the Patchwork Package in R
Просмотров 368Год назад
Working with the Patchwork Package in R
Gonzalez Anthony Young Jason Jones William
Smith Susan Rodriguez Anna Williams John
Jackson Maria White Daniel Brown Edward
Thomas Helen Rodriguez Jennifer Lewis Robert
Please share dataset and code
Results from this experiment would be very helpful. I hope it works.
Thank you for your great teaching!
Am not able to launch project(6) . Is there a way for first time users to do it
Thank you for your question. This seems to be working fine on my end. Could you share what error message R is giving you? Are you able to launch other projects?
This study assumes that all papers are equal because they are academic in nature. This can't be farther from the truth. Heterodox journals publish different types of papers than mainstream journals do because heterodox tools and methods are different. The neoclassical economics articles are within the dominant paradigm and commonly econometrics/optimization studies of the problems from that perspective. They are not philosophical or reasoned articles; they don't have any historical approach; they don't look at other paradigms, even if they don't know that they exist. In contrast, the heterodox paradigms constantly use approaches that require this boundary. They need to define themselves, formulate, and justify their perspectives in contrast to the flawed dominant paradigm. This is a very good example of data science insufficiency in tackling a problem. While plotting a few time series is useful, it is rarely enough to understand a phenomenon. Answering a question such as "Is neoclassical economics a good term?" requires matching the neoclassical definitions and distinctive principles with what appears in their academic journals and textbooks. It is not a simple data science question. Nevertheless, data science can help. This study only demonstrates that mainstream journal articles use the term less frequently. Nothing more.
I wish the videos were run in order
sql works same but more clarity in R
Can't understand nested code
thank you
Thank you for simplifying this. I subscribed immediately. It seems nobody really knew how to simplify it
The best explanation i have watched on R and R studio, thanks
This was really helpful , Thank you !
What you came here for starts at 0:45
Unbeliavable bullshitter!! R is much much more than a pipe-operator!!!!!
Thanks for the video. Kindly share the dataset
super helpful!
Thank you! Learning pipes initially was a bit terrifying, but you made it so easy,😀 and completely understand now. On to the next lesson.
Underrated explanation, I was taking a course and was still confused. You cleared it up quick.
Could you please provide us the dft_df.csv file
This video is not accurate. install.packages(“tidyverse”) downloads and installs ALL of the tidyverse packages. However, library(tidyverse) only loads the core 9 packages.
which link below?
very good ... can you send the data after preprocessing ... and the preprocessing code
Hello Good day. Could I get access to your data and methods. I'm trying to do something similar in terms of measuring productivity but im not well versed in R. I could really appreciate if you could provide a link/source to your studies and how you went about trying to solve this. thanks
where is the dataset
hey, did you proceed with this code? do you have the dataset?
the Tidyverse is infinitely superior for DS than Python.
very powerful and simple video
Where is the link?
I’m in my DS masters programme and this helped me understand pipe so much! Thank you!
great video
hello its really hellpful video. i am working with limma but I am having problem with the ebayes() function. please help me with it. and also can you provide the script of this tutorial.
I’m a little surprised because if you’re able to run lmFit() without any errors, then you should also be able to run efit(). What error are you getting? Thanks for watching!
thank you so much for explaining this like Im 5 y/o
💃 'Promo sm'
That was helpful and good, kindly provide the dataset and github link friend
hey, did you proceed with this code? do you have the dataset?
Interesting use case of computation and programming.
Luffy hat?
Thank you for making this review. Much appreciated.
Mine - for a book project in Quarto where do you suggest the libraries should be loaded? In each chapter separately or in some sort of front end document? As a LaTeX user I would be apt to put all my libraries (in LaTeX usepackage) in a single place. Does it matter?
Thank you.
Thank you very much. This is straight forward and very clearly explained.
Hey I am trying to run the single cell experiment, umi <- SingleCellExperiment(assays = list(counts = as.matrix(molecules)), colData = annotation), but I am getting this error, Error in validObject(.Object) : invalid class "SummarizedExperiment" object: nb of cols in 'assay' (5) must equal nb of rows in 'colData' (19027). Why is this happening and how do I fix it? Thank you
thank you
Thx so much for that lecture! You help me a lot!
Wakapenga iwewe 🔥🔥🔥
Mwana wevhu
Hoyo
Such a good explanation! Can you help me. I keep getting this error-Error in `bind_rows()`: ! Can't combine `..1$started_at` <datetime<UTC>> and `..2$started_at` <character>. Backtrace: 1. dplyr::bind_rows(q2_2019, q3_2019, q4_2019, q1_2020) . U have tried everything. I apologize but I'm really new to this. Thank you in advance.
Thank you so much. As a beginner, these exercises are helpful.