![Galit Shmueli](/img/default-banner.jpg)
- Видео 101
- Просмотров 823 559
Galit Shmueli
Добавлен 11 окт 2011
Lecture videos on Data Mining, Predictive Analytics and Business Intelligence
BADM Project Presentations: InterAgent
Two final projects in the Business Analytics Using Data Mining course (NTHU, Institute of Service Science), Fall 2022. Collaboration with InterAgent on predicting factory delivery time.
Просмотров: 154
Видео
Prediction in Causal Research: Part 1/2
Просмотров 419Год назад
In this two-part lecture (as part of the PhD Research Methods course at National Tsing Hua University's Institute of Service Science), we look the important role prediction can play in scientific theory-building, as a complement to causal studies. Part 1 provides the background and focuses on the differences between causal explanation, prediction, and description. Key paper: Shmueli, G. (2010),...
Prediction in Causal Research: Part 2/2
Просмотров 172Год назад
[Lecture as part of the PhD Research Methods course at National Tsing Hua University's Institute of Service Science] In this second part, I discuss how prediction can be used for enhancing theory-building causal research and what has changed in recent years in top journals in terms of injecting prediction into causal research. Relevant papers: 1. Shmueli G., and Koppius, O. R. (2011), Predictiv...
2021 Gosset Lecture: The Language of Statistics (and What's Lost in Translation)
Просмотров 6702 года назад
Galit Shmueli from National Tsing Hua University presents the ISBIS Gosset lecture at the 63rd ISI World Statistics Congress, titled "The Language of Statistics (and What's Lost in Translation.)" www.isbis-isi.org/gosset.html Anstract The field of statistics uses “statistical language” for describing phenomena in the observable world. This language includes statistical notation for parameters, ...
"Improving" Prediction using Behavior Modification (Keynote at ISF 2021)
Просмотров 4072 года назад
International Journal of Forecasting Editor’s Invited Talk at the 41st International Symposium on Forecasting (ISF21)
Yoga Session for 2020 International Conference on Information Systems (ICIS)
Просмотров 7713 года назад
A simple, traditional yoga session for all levels (even first timers). Grab a yoga mat and a hard cushion or folded towel and join us for 24 minutes of body and mind rejuvenation! Instructed by Galit Shmueli. Produced for 2020 International Conference on Information Systems (ICIS). Thanks to students Kellan Nguyen and Fan-Hao Huang from NTHU Institute of Service Science. Produced by the NTHU Of...
Introducing the IJDS for 2020 INFORMS DM+QSR Editors panel
Просмотров 3123 года назад
This video provides a short introduction by EIC Galit Shmueli of the newly launched INFORMS Journal on Data Science. The presentation took place at the 2020 INFORMS panel on "Editor's Perspective in Publishing Data Science-Focused Papers". (This session was jointly sponsored by the QSR and Data Mining communities.) www.abstractsonline.com/pp8/#!/9022/session/3316
Introducing the IJDS at the 2020 INFORMS joint Joint DMDA/DS Editorial Panel Session
Просмотров 1823 года назад
This video provides a short introduction by EIC Galit Shmueli of the newly launched INFORMS Journal on Data Science. The presentation took place at the 2020 INFORMS joint Editorial Panel Session of the DMDA and Data Science Workshops (Nov 7, 2020). www.abstractsonline.com/pp8/#!/9022/session/3503
2020 Teaching Innovation Award by INFORMS Information Systems Society: Award Speech by Galit Shmueli
Просмотров 3123 года назад
My short video for receiving the 2020 inaugural INFORMS Information Systems Society Teaching Innovation Award connect.informs.org/iss/awards/award-winners
"Improving" Prediction of Human Behavior using Behavior Modification (ENBIS-20)
Просмотров 6553 года назад
ENBIS-20 Online Conference Invited Talk by Galit Shmueli
ACEMS Public Lecture: "Improving" Prediction of Human Behavior Using Behavior Modification
Просмотров 5843 года назад
In this ACEMS Virtual Public Lecture, Prof Shmueli discusses how internet platforms can combine prediction with behavior modification to "improve" prediction of human behavior. Event info and abstract: acems.org.au/events/acems-virtual-public-lecture-improving-prediction-human-behavior-using-behavior-modification See paper here arxiv.org/abs/2008.12138
Wrestling Prediction Error: Better Predictions or Better "Actuals"?
Просмотров 4483 года назад
Talk at SCECR Symposium 2020 by Galit Shmueli - I examine the combination of predictive models and behavior modification, as can be used by internet platforms to create better "prediction products". - I introduce the use of Judea Pearl's do() operator to allow combining predictive and intervention modeling, and then I analyze the effect and implications of the above strategy.
Reinventing the Data Analytics Classroom
Просмотров 9545 лет назад
Keynote at "International Symposium on University Teaching Innovation", National Yunlin University of Science & Technology, Taiwan, May 2019 Slides for the talk: www.slideshare.net/gshmueli/reinventing-the-data-analytics-classroom
BADM 5.3 Hierarchical Clustering Part 2
Просмотров 2,1 тыс.6 лет назад
Hierarchical Clustering Part 2 of 2 This video was created by Professor Galit Shmueli and has been used as part of blended and online courses on Business Analytics using Data Mining. It is part of a series of 37 videos, all of which are available on RUclips. For more information: www.dataminingbook.com gshmueli dataminingbook Here is the complete list of the videos: • W...
Discriminant Analysis: Misclassification costs and over-sampling (Part 3)
Просмотров 7 тыс.6 лет назад
Misclassification costs; Over-sampling and "distorted" samples; Average cost per classified record; Incorporating external class probabilities into classification functions; Incorporating misclassification costs into classification functions; Effect of external class probabilities and misclassification costs on classifying new records This video was created by Professor Galit Shmueli and has be...
Discriminant Analysis: Statistical Distance (Part 2)
Просмотров 8 тыс.6 лет назад
Discriminant Analysis: Statistical Distance (Part 2)
BADM 9.2 Logistic Regression for Classification
Просмотров 3,4 тыс.6 лет назад
BADM 9.2 Logistic Regression for Classification
BADM 9.1 Logistic Regression for Profiling
Просмотров 2,5 тыс.6 лет назад
BADM 9.1 Logistic Regression for Profiling
BADM 8.3 Classification and Regression Trees Part 3
Просмотров 1,6 тыс.6 лет назад
BADM 8.3 Classification and Regression Trees Part 3
BADM 8.2 Classification and Regression Trees Part 2
Просмотров 2,9 тыс.6 лет назад
BADM 8.2 Classification and Regression Trees Part 2
BADM 8.1 Classification and Regression Trees Part 1
Просмотров 4 тыс.6 лет назад
BADM 8.1 Classification and Regression Trees Part 1
I really love your explanation. God bless u mam
Great video! But what if the seasonal period is unknow or time-varying? how can i estimate it?
By Lt and Lt-1, I assume this is the same as notation in your book that uses the Ft and Ft-1 for the forecast value?
Thanks for the video. I want to predict the value of one Index for 2019 using past data from 1970-2018. This Index is a TS, but it is also calculated using variables with weights that also are TS. What to do? To use only the data for I(t) like in 10:30 or to use the variables? Honestly I try by using SPSS where for Input I use the Variables data from 1970-2018, and for Output data I mark only the Index. It can show me only the correlation between Variables and the Index with synaptic weights but nothing more. I cannot find the forecasting option for 2019, I can only plot the Predicted values for the Index from 1970-2018, but cannot go further. What to do in order to predict the Index for 2019? I want to use ANN for the prediction but cannot find a way to do it. Thanks!
Thank you. This was helpful
Thanks, this is a good explanation of roll-forward partitioning.
Thank you very much for the excellent moving average method class. It was very useful for me. I´d appreciate more videos on Forecasts and, if possible, in DOE too.
You must work on fixing those that wish to tell others what to do before you worry about making it easier to get people to follow along. We have evil people in the world already abusing so much. This is inhumane as of now and the trajectory is dire and unbearable
This is a sickening path not worth living in
nicely done
Amazingly explained. Clear, to the point, brief. Thank you!
Superd ❤
Thank you for the clear explanation on these forecasting models!!!
What about the assumption of independent observations in logistic regression model?
Can we combine exponential smoothing with regression?
Sir, If I am doing double exponential smoothing 5 period moving average with a software program--after 5 period exponential moving average is calculated say (X) does computer do second calculation with data X and (X-t1) (X-t2) (X-t3) (X-t4)- - - - forgive me,I went to college 50 years ago..(X-t1) is exponential moving average one period prior & so on.
What is exact meaning of level?
As a beginner learning time series analysis and forecasting, I think your lesson is much clearer and better than what I've seen so far. Thank you for sharing and teaching! :)
Interesting view on missing values!
Thanks for the nice presentation. What software (GUI) you have used? Please provide a link to it, if possible.
Very well explained. I encourage you to do more educational videos!
I love it!
Amazing video!!!
It would be nice if you touched on how to add back the trend or seasonality.
God of data science 🙏🏼
Amazing 🤩
This is a great course from a great professor, Dr. Galit Shmueli. I really liked your teaching way. Everything is so clear. Thank you so much.
Do feel RUclips is much better than the professor in uni..
At 3:35 L(t) = ... should be L(t+1) = ... .Do you agree? (nice lecture by the way)
Thank you so much
Can you hire a presenter to speak and make the presentation more engaging? I can't stay awake.
kol hakavod Galit :)
Excellent , much appreciated. Thank you very much
I have a question - What is the difference between autocorrelation and seasonality. Can I say, "Presence of seasonality at time instances of lag distance apart is found by measuring autocorrelation behaviour mathematically" ? Thank you
I don't have words to praise your content. I am a self learner from this kind of knowledge transfer. The first person I admire was Andrew NG and you are added to that Bubble. I appreciate the selection of real life examples. I have been through similar lectures which generates the stationary time series and adds trend and seasonality then demonstrates how to identify, measure, detach these components. That treatment made me feel what's going on but I felt like it is artificial and it appeared what we have added we are identifying. I loved your comparisons viz., - comparison among the different methods with gradual increment in complexity - comparison time series problems with other machine learning problems In summary, it is the best course I have ever come across !! Thank you Anandachetan Elikapati
although back in 2016 , but super explanation. sharp and crystal !
super useful, ty!
1:34
How can I fill the missing data in matlab?
Very good!
3:00 Why might a data analyst want to remove seasonality if all we need to do is remove the trend to make the data stationary?
You are so amazing in the way you presented that... Thank God... I'm unsure how i found your page, but I'm hoping to follow plenty more
if the result of the matrix multiplication is negative then what do we do. Do we take the absolute value?
Awesome
Fantastic stuff. Put words to splinter I couldn't extract from my brain until now.
This is the best video I have seen on the ARIMA model so far. GREAT Job!
I noticed that in the matrix example, the order of the variables of the individual (CCAv, Age, Income) isnt the same that te order of the variables on the covariance matrix (Income, AAv,Age). Please chek it
You are a fucking treasure
Easy explained maybe complex problematics. I understand, thank you!
Much appreciated.