Predicting non-state terrorism worldwide

Python, André; Bender, Andreas; Nandi, A.; Hancock, Penelope A.; Arambepola, Rohan; Brandsch, Jürgen; Lucas, Tim

doi:10.1126/sciadv.abg4778

Cited by 17 publications

(6 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…XGBoost is based on a gradient‐boosting decision tree method (Friedman, 2001) and has been recently applied in a wide range of applications aiming to predict complex spatial phenomena at the global scale (e.g., Cook‐Patton et al., 2020; Python et al., 2021; Zheng et al., 2021). XGBoost uses a gradient‐descent algorithm to minimize the loss when adding new models.…”

Section: Methodsmentioning

confidence: 99%

Machine Learning Investigation of Clinopyroxene Compositions to Evaluate and Predict Mantle Metasomatism Worldwide

Qin

Huang

et al. 2022

JGR Solid Earth

Self Cite

View full text Add to dashboard Cite

Plain Language Summary Clinopyroxene is a major mineral in Earth's upper mantle. Previous studies have attempted to discriminate between reactions modifying the mantle by plotting clinopyroxene major and trace element compositions in two‐dimensional (2‐D) diagrams. However, these 2‐D methods show poor accuracy when applied to global datasets. Therefore, we suggest a machine learning approach to evaluate clinopyroxene compositional data in higher dimensions. Our results demonstrate that machine learning can significantly improve the accuracy of clinopyroxene compositional predictions over classical methods utilizing elemental ratios. Furthermore, the application of our algorithm to a global clinopyroxene dataset suggests that mantle metasomatism is globally widespread.

show abstract

Section: Methodsmentioning

confidence: 99%

Machine Learning Investigation of Clinopyroxene Compositions to Evaluate and Predict Mantle Metasomatism Worldwide

Qin

Huang

et al. 2022

JGR Solid Earth

Self Cite

View full text Add to dashboard Cite

show abstract

“…Additionally, our task is to predict terrorist attacks or events, which is distinct from works that have sought to infer characteristics of an attack, such as the responsible group, after it has taken place [2,5,31,32]. Few works have attempted to predictively model terrorist attacks, with the recent work of Python et al being the noteworthy exception [25]. In this study, the authors train several machine learning models using prior terrorism data (from the GTD) in conjunction with other geographic and socioeconomic features to predict attacks at discrete spatiotemporal intervals.…”

Section: Introductionmentioning

confidence: 99%

Predicting Terrorist Attacks in the United States using Localized News Data

Krieg,

Smith,

Chatterjee

et al. 2022

Preprint

View full text Add to dashboard Cite

Terrorism is a major problem worldwide, causing thousands of fatalities and billions of dollars in damage every year. Toward the end of better understanding and mitigating attacks that take place in the United States, we present a set of machine learning models that learn from localized news data in order to predict whether a terrorist attack will occur on a given calendar date and in a given state. The best model-a Random Forest that learns from a novel variable-length moving average representation of the feature spaceachieves area under the receiver operating characteristic scores > .667 on four of the five states that were impacted most by terrorism between 2015 and 2018. Our key findings include that modeling terrorism as a set of independent events, rather than as a continuous process, is a fruitful approach-especially when the events are sparse and dissimilar. Additionally, our results highlight the need for localized models that account for differences between locations. From a machine learning perspective, we found that the Random Forest model outperformed several deep models on our multimodal, noisy, and imbalanced data set, thus demonstrating the efficacy of our feature representation method in such a context. Finally, we analyze factors that limit model performance, which include a noisy feature space and small amount of available data. These contributions provide an important foundation for the use of machine learning in efforts against terrorism in the United States and beyond.

show abstract

“…Additionally, our task is to predict the occurrence of attacks or events, which is distinct from works that have sought to infer characteristics of an attack, such as the responsible group, after it has taken place [ 5 – 9 ]. Few works have attempted to predictively model terrorist attacks, with the recent work of Python et al being one exception [ 10 ]. The authors trained several machine learning models using prior terrorism data (from the GTD) in conjunction with other geographic and socioeconomic features to predict attacks at discrete spatiotemporal intervals.…”

Section: Introductionmentioning

confidence: 99%

“…In order to make meaningful predictions at a granular timescale (e.g., will a terrorist attack take place during a given week), models must have inputs of a similar temporal granularity in order to differentiate between points in time. Many of the inputs utilized by Python et al in [ 10 ] are relatively static, such as population density and gross domestic product (GDP). The only temporally granular inputs were autoregressions based on local terrorist activity (i.e., whether there were recent terrorist attacks in the location of concern).…”

Section: Introductionmentioning

confidence: 99%

Predicting terrorist attacks in the United States using localized news data

et al. 2022

View full text Add to dashboard Cite

Terrorism is a major problem worldwide, causing thousands of fatalities and billions of dollars in damage every year. To address this threat, we propose a novel feature representation method and evaluate machine learning models that learn from localized news data in order to predict whether a terrorist attack will occur on a given calendar date and in a given state. The best model (a Random Forest aided by a novel variable-length moving average method) achieved area under the receiver operating characteristic (AUROC) of ≥ 0.667 (statistically significant w.r.t. random guessing with p ≤ .0001) on four of the five states that were impacted most by terrorism between 2015 and 2018. These results demonstrate that treating terrorism as a set of independent events, rather than as a continuous process, is a fruitful approach—especially when historical events are sparse and dissimilar—and that large-scale news data contains information that is useful for terrorism prediction. Our analysis also suggests that predictive models should be localized (i.e., state models should be independently designed, trained, and evaluated) and that the characteristics of individual attacks (e.g., responsible group or weapon type) were not correlated with prediction success. These contributions provide a foundation for the use of machine learning in efforts against terrorism in the United States and beyond.

show abstract

Predicting non-state terrorism worldwide

Cited by 17 publications

References 79 publications

Machine Learning Investigation of Clinopyroxene Compositions to Evaluate and Predict Mantle Metasomatism Worldwide

Machine Learning Investigation of Clinopyroxene Compositions to Evaluate and Predict Mantle Metasomatism Worldwide

Predicting Terrorist Attacks in the United States using Localized News Data

Predicting terrorist attacks in the United States using localized news data

Contact Info

Product

Resources

About