| 1. |
- Bolin, David, 1983, et al.
(författare)
-
Functional ANOVA modelling of pedestrian counts on streets in three European cities
- 2021
-
Ingår i: Journal of the Royal Statistical Society. Series A, (Statistics in Society). - : Oxford University Press (OUP). - 1467-985X .- 0964-1998. ; 184:4, s. 1176-1198
-
Tidskriftsartikel (refereegranskat)abstract
- The relation between pedestrian flows, the structure of the city and the street network is of central interest in urban research. However, studies of this have traditionally been based on small data sets and simplistic statistical methods. Because of a recent large-scale cross-country pedestrian survey, there is now enough data available to study this in greater detail than before, using modern statistical methods. We propose a functional ANOVA model to explain how the pedestrian flow for a street varies over the day based on its density type, describing the nearby buildings, and street type, describing its role in the city’s overall street network. The model is formulated and estimated in a Bayesian framework using hour-by-hour pedestrian counts from the three European cities, Amsterdam, London and Stockholm. To assess the predictive power of the model, which could be of interest when building new neighbourhoods, it is compared with four common methods from machine learning, including neural networks and random forests. The results indicate that this model works well but that there is room for improvement in capturing the variability in the data, especially between cities.
|
|
| 2. |
- Hartvigsson, Elias, 1986, et al.
(författare)
-
Local forecasts of electrc vehicles for grid planning purposes
- 2022
-
Ingår i: IET Conference Proceedings. - : Institution of Engineering and Technology. - 2732-4494. ; 2022:3, s. 878-882
-
Tidskriftsartikel (refereegranskat)abstract
- Electrification of passenger vehicles is rapidly becoming the main alternative for decarbonizing transportation. The high power associated with charging of electric vehicles is likely to require actions from grid operators. Using machine learning and GIS analysis we produce forecasts of electric vehicles in very small cells, down to a few hundred meters for Norway. Using a baseline comparison, we find that a random forest model produces the overall lowest error, with a Mean Absolute Error of 14.0, and Mean Absolute Percentage Error of 33.9%. We find that both the existing vehicle fleet, and forecast shows that there is a large variation in electric vehicle adoption between cells. With knowledge where and when electric vehicles are adopted, grid operators can better plan their future investments related to electric vehicle charging, and thereby reduce investment costs.
|
|
| 3. |
- Horn, Christian, 1978, et al.
(författare)
-
Artificial Intelligence, 3D Documentation, and Rock Art—Approaching and Reflecting on the Automation of Identification and Classification of Rock Art Images
- 2022
-
Ingår i: Journal of archaeological method and theory. - : Springer Science and Business Media LLC. - 1072-5369 .- 1573-7764. ; 29, s. 188-213
-
Tidskriftsartikel (refereegranskat)abstract
- Rock art carvings, which are best described as petroglyphs, were produced by removing parts of the rock surface to create a negative relief. This tradition was particularly strong during the Nordic Bronze Age (1700–550 BC) in southern Scandinavia with over 20,000 boats and thousands of humans, animals, wagons, etc. This vivid and highly engaging material provides quantitative data of high potential to understand Bronze Age social structures and ideologies. The ability to provide the technically best possible documentation and to automate identification and classification of images would help to take full advantage of the research potential of petroglyphs in southern Scandinavia and elsewhere. We, therefore, attempted to train a model that locates and classifies image objects using faster region-based convolutional neural network (Faster-RCNN) based on data produced by a novel method to improve visualizing the content of 3D documentations. A newly created layer of 3D rock art documentation provides the best data currently available and has reduced inscribed bias compared to older methods. Several models were trained based on input images annotated with bounding boxes produced with different parameters to find the best solution. The data included 4305 individual images in 408 scans of rock art sites. To enhance the models and enrich the training data, we used data augmentation and transfer learning. The successful models perform exceptionally well on boats and circles, as well as with human figures and wheels. This work was an interdisciplinary undertaking which led to important reflections about archaeology, digital humanities, and artificial intelligence. The reflections and the success represented by the trained models open novel avenues for future research on rock art.
|
|
| 4. |
- Ivarsson, Oscar, 1988, et al.
(författare)
-
Towards Machine Learning on Data from Professional Cyclists
- 2018
-
Ingår i: Proceedings of the XII World Congress of Performance Analysis of Sport. ; 2018
-
Konferensbidrag (refereegranskat)abstract
- Professional sports are developing towards increasingly scientific training methods with increasing amounts of data being collected from laboratory tests, training sessions and competitions. In cycling, it is standard to equip bicycles with small computers recording data from sensors such as power-meters, in addition to heart-rate, speed, altitude etc. Recently, machine learning techniques have provided huge success in a wide variety of areas where large amounts of data (“big data”) is available. In this paper, we perform a pilot experiment on machine learning to model physical response in elite cyclists. As a first experiment, we show that it is possible to train a LSTM machine learning algorithm to predict the heart-rate response of a cyclist during a training session. This work is a promising first step towards developing more elaborate models based on big data and machine learning to capture performance aspects of athletes.
|
|
| 5. |
- Mao, Wengang, 1980, et al.
(författare)
-
Spatio-temporal modelling of wind speed variation
- 2018
-
Ingår i: Proceedings of the International Offshore and Polar Engineering Conference. - : International Society of Offshore and Polar Engineers. - 1098-6189 .- 1555-1792. ; 2018-June, s. 397-402
-
Konferensbidrag (refereegranskat)abstract
- Wind speed variability in the Northern North Atlantic has been success- fully modelled by a spatio-temporal transformed Gaussian field in our previous study. It was shown that this type of model does not describe correctly the extreme wind speeds attributed to tropical storms and hurri- canes. This spatio-temporal model was generalized to include the possi- bility of the occurrence of rare severe storms. In that work, the daily wind speed variability was modelled by the transformed Gaussian field, and then random components were added to model rare events with extreme wind speeds. The model was termed the hybrid model. The transformed Gaussian and the hybrid models are locally stationary and homogeneous random fields with localized dependence structure, which is described by time and space dependent parameters with a natural physical interpreta- tion. In the present study, these models are used to describe the variability of wind speed in other areas, i.e., the Caribbean sea, the South China Sea and the Arctic area. In most locations, the transformed Gaussian field is a sufficiently accurate model. However, in some regions, e.g. Laptev and the Beaufort Sea at the Arctic, this model severely underestimates the frequencies of extreme winds. In this study, the hybrid model is used to describe the wind variation in these regions. There are also locations, e.g. along the east coast of Greenland, most of the coast areas of the South China Sea, where frequencies of high wind speeds are severely overestimated by the transformed Gaussian model. In this paper, the models are fitted to ERA-Interim reanalysis wind data and used to find long-term distributions of wind speed, to estimate wind speed return values, e.g. 100-year extreme wind speed, and to compute the expected yearly frequency of events that wind speed exceeds a fixed threshold value.
|
|
| 6. |
- Sokolova, Ekaterina, 1986, et al.
(författare)
-
Data-driven models for predicting microbial water quality in the drinking water source using E. coli monitoring and hydrometeorological data
- 2022
-
Ingår i: Science of the Total Environment. - : Elsevier BV. - 0048-9697 .- 1879-1026. ; 802
-
Tidskriftsartikel (refereegranskat)abstract
- Rapid changes in microbial water quality in surface waters pose challenges for production of safe drinking water. If not treated to an acceptable level, microbial pathogens present in the drinking water can result in severe consequences for public health. The aim of this paper was to evaluate the suitability of data-driven models of different complexity for predicting the concentrations of E. coli in the river Göta älv at the water intake of the drinking water treatment plant in Gothenburg, Sweden. The objectives were to (i) assess how the complexity of the model affects the model performance; and (ii) identify relevant factors and assess their effect as predictors of E. coli levels. To forecast E. coli levels one day ahead, the data on laboratory measurements of E. coli and total coliforms, Colifast measurements of E. coli, water temperature, turbidity, precipitation, and water flow were used. The baseline approaches included Exponential Smoothing and ARIMA (Autoregressive Integrated Moving Average), which are commonly used univariate methods, and a naive baseline that used the previous observed value as its next prediction. Also, models common in the machine learning domain were included: LASSO (Least Absolute Shrinkage and Selection Operator) Regression and Random Forest, and a tool for optimising machine learning pipelines – TPOT (Tree-based Pipeline Optimization Tool). Also, a multivariate autoregressive model VAR (Vector Autoregression) was included. The models that included multiple predictors performed better than univariate models. Random Forest and TPOT resulted in higher performance but showed a tendency of overfitting. Water temperature, microbial concentrations upstream and at the water intake, and precipitation upstream were shown to be important predictors. Data-driven modelling enables water producers to interpret the measurements in the context of what concentrations can be expected based on the recent historic data, and thus identify unexplained deviations warranting further investigation of their origin.
|
|
