Name
Abstract | ||
|---|---|---|
Being able to predict the daily activity of firefighters is of great interest to optimize human and material resources. It will allow to enable a quicker response by achieving a better geographical deployment of these resources according to the expected number of interventions. Having obtained the list of interventions for the period 2012-2017 in the Department of the Doubs, France, we added a relevant collection of explanatory variables based on calendar data (time of day, day of the week, day of the month, year, public holidays, etc.), road traffic, meteorological and astronomical data, and so on. After detecting outliers and completing missing data, this set has been divided for learning, validating, and testing. The learning is then carried out on an ad hoc multilayer perceptron whose hyperparameters are finely defined using some super-computer facilities. This neural architecture are finally applied on a real case study, that is, to the predictions of firemen interventions for the year 2017 after a learning stage on 2012-2016, leading to really encouraging results. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1007/978-3-030-29516-5_42 | INTELLIGENT SYSTEMS AND APPLICATIONS, VOL 1 |
Keywords | DocType | Volume |
Prediction algorithm, Artificial intelligence, Machine learning, Multilayer perceptron, Deep neural network, Incomplete dataset, Firemen interventions | Conference | 1037 |
ISSN | Citations | PageRank |
2194-5357 | 0 | 0.34 |
References | Authors | |
0 | 7 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Christophe Guyeux | 1 | 332 | 61.61 |
| Jean-Marc Nicod | 2 | 0 | 0.34 |
| Christophe Varnier | 3 | 71 | 9.09 |
| Zeina Al Masry | 4 | 0 | 0.34 |
| Nourredine Zerhouny | 5 | 0 | 0.34 |
| Nabil Omri | 6 | 0 | 0.34 |
| Guillaume Royer | 7 | 1 | 1.71 |