Поиск
  • Поиск
  • Мои Раскадровки

CW1

Создать Раскадровку
Скопируйте эту раскадровку
CW1
Storyboard That

Создайте свою собственную раскадровку

Попробуйте бесплатно!

Создайте свою собственную раскадровку

Попробуйте бесплатно!

Текст Раскадровки

  • Concept
  • An Immune-system inspired technique to solve a real world big data problem of identifying incident hot spots of heavy goods vehicles(HGVs).
  • Implementation
  • SeleSup HSID algorithm - which groups similar instances together called clusters and identifies its centres (hot spots) by proximity measurement. Once the cluster centres are established, the rest of the data points are excluded.The authors used parallel implementation of SeleSup HSID built on Apache Spark to handle large data.
  • Core Related Work
  • This paper builds on the previous work of G. P. Figueredo et al (2012) which is about an immune-inspired instance selection mechanism for supervised classification. The related work section discusses several commonly used methods for identifying hot spots and their limitations with respect to analyzing large volumes of data.
  • Data Characteristics
  • The authors investigate a large telematics data set of 1,000,612 HGV incidents which occurred over a period of three-months in 2015 in the UK.The distribution is 773,323 speeding, 213,697 harsh braking and 13,592 harsh cornering incidents.Spatial Data: Location of the incident.Temporal Data: Date and time of the incident.Categorical Data: Type of the incident (speeding, harsh braking , harsh cornering).
  • Type of Evaluation
  • The technique was applied to four real-world datasets to evaluate the accuracy and effectiveness of the proposed method.The authors examined two different scenarios with constraints, including mileage limit, course, and address. They investigated the impact on the results across multiple independent runs, the effect of the initial size of the suppressor set, and how the mileage range influences the final number of hot spots.The results have been confirmed and validated through various interactions with transportation experts.
  • Representative Images
  • Image Attributions:590022 (https://www.pexels.com/photo/chart-close-up-data-desk-590022/) - Lukas - License: Free To Use / No Attribution Required / See https://www.pexels.com/license/ for what is not allowed2008478 (https://pixabay.com/vectors/diagram-icon-business-symbol-chart-2008478/) - Memed_Nurrohmad - License: Free for Most Commercial Use / No Attribution Required / See https://pixabay.com/service/license/ for what is not allowed

Image Attributions

  • 2008478 - Memed_Nurrohmad - (Лицензия Free for Most Commercial Use / No Attribution Required / See https://pixabay.com/service/license/ for what is not allowed )
  • 590022 - Lukas - (Лицензия Free To Use / No Attribution Required / See https://www.pexels.com/license/ for what is not allowed )
Создано более 30 миллионов раскадровок