filmov
tv
Forecasting airline passengers using designer machine learning - Alexander Backus, Jan van der Vegt
Показать описание
PyData Amsterdam 2018
The ability to accurately forecast the amount of passengers that will board a particular flight is crucial for airline operations. But how do we design a machine learning algorithm for this use case and in what ways can we improve it? In this talk, we start with a simple linear model, evolving to increasingly complex deep learning neural network architectures.
--
PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.
PyData conferences aim to be accessible and community-driven, with novice to advanced level presentations. PyData tutorials and talks bring attendees the latest project features along with cutting-edge use cases.
0:00 Introduction
0:50 Problem: Predicting Passenger Number & Use Cases
1:38 Problem: Unique Forecasting Constraint - Shrinking Horizon
2:08 System Requirements
2:50 System Design
3:23 "Designer Machine Learning" Definition
3:56 Data: Artificial Flight-bookings
5:07 Data: Features
6:24 Model: Simple Linear Model & ANN
7:15 Model: Feed-Forward Deep Neural Network
7:48 Model: Loss Function - MSE
8:25 Keras Code Example
9:33 Use Case: Aircraft Allocation
10:25 Evaluation: Probability of Capacity Overflow
11:45 Model: Conditional Density Estimation
13:48 Model: Updated ANN Outputs (Mu & Sigma) & Loss Function
15:04 Keras Code Example for Conditional Density Estimation
16:46 Model: Mixture Density
17:09 Model: Mixture Density Networks
17:51 Challenges: Selecting Distributions & Numerical Optimization
19:06 Sequence Feature Extraction
19:41 Model: Representational Learning & Recurrent Neural Network
21:08 Keras Code Example for RNN with LSTM
21:57 Challenges: Non-uniform Time Deltas & Flight Dependencies
23:08 Key Take-aways
24:32 Q&A: Q1
26:04 Q&A: Q2
27:26 Q&A: Q3
29:16 Q&A: Q4
30:58 Q&A: Q5
32:40 Q&A: Q6
The ability to accurately forecast the amount of passengers that will board a particular flight is crucial for airline operations. But how do we design a machine learning algorithm for this use case and in what ways can we improve it? In this talk, we start with a simple linear model, evolving to increasingly complex deep learning neural network architectures.
--
PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.
PyData conferences aim to be accessible and community-driven, with novice to advanced level presentations. PyData tutorials and talks bring attendees the latest project features along with cutting-edge use cases.
0:00 Introduction
0:50 Problem: Predicting Passenger Number & Use Cases
1:38 Problem: Unique Forecasting Constraint - Shrinking Horizon
2:08 System Requirements
2:50 System Design
3:23 "Designer Machine Learning" Definition
3:56 Data: Artificial Flight-bookings
5:07 Data: Features
6:24 Model: Simple Linear Model & ANN
7:15 Model: Feed-Forward Deep Neural Network
7:48 Model: Loss Function - MSE
8:25 Keras Code Example
9:33 Use Case: Aircraft Allocation
10:25 Evaluation: Probability of Capacity Overflow
11:45 Model: Conditional Density Estimation
13:48 Model: Updated ANN Outputs (Mu & Sigma) & Loss Function
15:04 Keras Code Example for Conditional Density Estimation
16:46 Model: Mixture Density
17:09 Model: Mixture Density Networks
17:51 Challenges: Selecting Distributions & Numerical Optimization
19:06 Sequence Feature Extraction
19:41 Model: Representational Learning & Recurrent Neural Network
21:08 Keras Code Example for RNN with LSTM
21:57 Challenges: Non-uniform Time Deltas & Flight Dependencies
23:08 Key Take-aways
24:32 Q&A: Q1
26:04 Q&A: Q2
27:26 Q&A: Q3
29:16 Q&A: Q4
30:58 Q&A: Q5
32:40 Q&A: Q6
0:33:59
Forecasting airline passengers using designer machine learning - Alexander Backus, Jan van der Vegt
1:03:50
Grushka-Cockayne: Forecasting Airport Transfer Passenger Flow Using Real-Time Data & Machine Sem...
0:45:05
TEQnation 2019: Alexander Backus & Daan Debie - Forecasting passengers with machine learning at ...
0:22:45
Keflavik Airport Case Study: Passenger Flow and Forecasting
0:01:41
IATA's 20 Year Passenger Forecast
0:25:31
Airline passenger referral prediction| Machine Learning Classification Full Project @AlmaBetter
0:06:36
American Airlines Passenger Prediction
0:01:07
20 Year Passenger Forecast 2016
0:01:57
Study finds airline passenger satisfaction is down
0:13:23
Student Dataset Presentation: Airlines Passenger Satisfaction Forecast
0:13:56
Time Series forecasting using ARIMA model | Air Passenger Dataset | part 2
0:08:14
Airline Passenger Satisfaction Part-1 | Intro| Machine Learning Project End to End implementation
0:11:30
Research on 'Passengers' satisfaction prediction in airline service'
0:15:01
ITS665 FINAL PROJECT: AIRLINE PASSENGER SATISFACTION
0:28:14
TensorFlow 2.0 Tutorial for Beginners 18 - Airlines Passenger Prediction using RNN - LSTM
0:45:37
Airline Passenger Satisfaction using XGBoost Boosting || Random Forest Bagging ||
0:00:13
🇧🇷 Shark Attack | Rio De Janeiro Atlantic Ocean #shark #travel #subscribe #comedy #brazil #ocean...
0:12:53
Flight Secrets That Are Never Told To Passengers
0:26:50
Episode 1: Passenger growth + infrastructure standstill = bottleneck?
0:07:59
SUBWAY PASSENGER FLOW FORECASTING WITH MULTI-STATION AND EXTERNAL FACTORS
0:02:01
Secret program watches, follows airline passengers
0:00:49
5 key findings from the 2017 Global Passenger survey
0:59:05
Air passenger webinar
0:00:16
Do you want to better your life? #philippines #angelescity #expat #pampanga #travelvlog
Комментарии