Loading...
Publication
Agregação de dados com a metáfora da Rosa dos Ventos
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 2.81 MB | Adobe PDF |
Authors
Abstract(s)
A rosa dos ventos é um tipo de gráfico radial muito usado em meteorologia para representar ventos dominantes num dado local, agregando múltiplos registos de velocidade e direção recolhidos ao longo de um período de tempo, por exemplo num ano. Neste trabalho, a rosa dos ventos é utilizada para representar dados agregados relacionados com trajetos pessoais, sendo que um trajeto é uma sequência de pontos geográficos que possuem uma data, hora e velocidade instantânea. Este projeto foi dividido em três objetivos: a criação de uma biblioteca que permite criar rosas dos ventos; a criação de uma aplicação que permite representar dados de trajetos pessoais; e testes com utilizadores. A biblioteca rosa dos ventos foi criada de forma a que haja bastante controlo sobre o desenho de uma rosa dos ventos e dos possíveis elementos circundantes da mesma. A aplicação rosa dos ventos, permite a visualização de trajetos num mapa e possibilita a interação com esses trajetos, estes são suscetíveis ao clique sendo esta a interação que despoleta a criação de um gráfico rosa dos ventos que representa informação relacionada com o ponto geográfico clicado. Assim, a rosa gerada mostra as direções dos trajetos que cruzam uma região circular centrada nesse ponto. A rosa também permite ver qual a origem e o destino dos trajetos, recorrendo a setas em redor do gráfico. É possível alterar o raio de agregação de uma rosa e o período de agregação mostrando, por exemplo, apenas trajetos efetuados de manhã. Finalmente, é possível ter mais que uma rosa dos ventos no ecrã, permitindo a comparação de gráficos. Foram realizados testes com utilizadores, tendo para tal sido criada uma aplicação para recolher dados da sua prestação. Os resultados indicam que os erros foram quase inexistentes e os tempos de resposta tiveram uma mediana de 18 segundos. Pode, portanto, considerar-se que a rosa dos ventos, no contexto de interpretação de trajetos num mapa, possui um nível de complexidade de leitura baixo, pelo que se considera que pode ser útil na análise de cenários semelhantes e no apoio à decisão.
The windrose is a radial chart that is widely used in meteorology to show the dominant winds in some place, by aggregating multiple registers of speed and direction recorded during some time period, e.g., a year. This project’s main idea is to use windroses in the representation of aggregated data associated with personal tracks, keeping in mind that a track is a set of points which have a date, hour and instant speed. The project was divided into three objectives: the creation of a library that enables the creation of windroses; the creation of an app that enables the representation of personal tracks; and user tests. The windrose library was created in order to have control of all drawing aspects of a windrose chart including the addition of elements surrounding the windrose. The windrose app enables the visualization of tracks in a map and also enables interaction with those tracks, meaning that, after the tracks have been drawn on a map, they are clickable, and when clicked in any point, a windrose is generated. This way, when a track is clicked, the generated windrose shows information regarding a circular area around that point defined by the user. The windroses generated by the click also allow to understand what is the origin of the whole tracks as well as their destination. It is also possible to change the aggregation radius and the aggregation period (periods of the day, like, morning, afternoon, evening and night). Ultimately, it is possible to have more than one windrose on screen, allowing for an easy comparison of data. To understand if the environment created by the app, i.e., the map and windroses are or not easy to comprehend, some user tests took place. It was created another app, which allows to get user performance data. Either individually or when in groups by age or gender, the results show that the errors were almost inexistent and the response times had a median of 18 seconds. This suggests that the windrose, in the context of interpretation of tracks in a map, has a low level of complexity, which makes the windrose a viable option for the analysis of similar scenarios and in decision making contexts.
The windrose is a radial chart that is widely used in meteorology to show the dominant winds in some place, by aggregating multiple registers of speed and direction recorded during some time period, e.g., a year. This project’s main idea is to use windroses in the representation of aggregated data associated with personal tracks, keeping in mind that a track is a set of points which have a date, hour and instant speed. The project was divided into three objectives: the creation of a library that enables the creation of windroses; the creation of an app that enables the representation of personal tracks; and user tests. The windrose library was created in order to have control of all drawing aspects of a windrose chart including the addition of elements surrounding the windrose. The windrose app enables the visualization of tracks in a map and also enables interaction with those tracks, meaning that, after the tracks have been drawn on a map, they are clickable, and when clicked in any point, a windrose is generated. This way, when a track is clicked, the generated windrose shows information regarding a circular area around that point defined by the user. The windroses generated by the click also allow to understand what is the origin of the whole tracks as well as their destination. It is also possible to change the aggregation radius and the aggregation period (periods of the day, like, morning, afternoon, evening and night). Ultimately, it is possible to have more than one windrose on screen, allowing for an easy comparison of data. To understand if the environment created by the app, i.e., the map and windroses are or not easy to comprehend, some user tests took place. It was created another app, which allows to get user performance data. Either individually or when in groups by age or gender, the results show that the errors were almost inexistent and the response times had a median of 18 seconds. This suggests that the windrose, in the context of interpretation of tracks in a map, has a low level of complexity, which makes the windrose a viable option for the analysis of similar scenarios and in decision making contexts.
Description
Trabalho de projeto de mestrado, Engenharia Informática (Sistemas de Informação )Universidade de Lisboa, Faculdade de Ciências, 2017
Keywords
Sistemas de informação Visualização analítica Rosa dos Ventos Trabalhos de projeto de mestrado - 2017