Sampling, Intervention, Prediction, Aggregation: A Generalized Framework for Model-Agnostic Interpretations

Scholbeck, Christian A.; Molnar, Christoph; Heumann, Christian; Bischl, Bernd; Casalicchio, Giuseppe

doi:10.1007/978-3-030-43823-4_18

Cited by 30 publications

(27 citation statements)

References 18 publications

(39 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…If this number is exceeded, production of waste tends to be higher. These findings are consistent with some studies that have reported a directly proportional relationship between the size of buildings and the generation of waste [ 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 ].…”

Section: Discussionsupporting

confidence: 93%

“…Other studies developed a general regression neural network (GRNN) model for the prediction of annual municipal solid waste (MSW) generation at the national level for 44 countries of different size, population and economic development level [ 9 ]. Deep learning model combined with permutation-based VI scores method (R package “vip”) [ 44 ] and individual conditional expectation (ICE) plots method [ 45 ] was the very innovative and modern approach of the study of this paper to forecast MSW independent variables, to determine the factors that most influence the variables and to estimate how much they affect them. In addition, MSW production, separate collection, and MSW management costs with the same database were considered in this study.…”

Section: Discussionmentioning

confidence: 99%

“… Identification of factors most influenced each of the three independent variables within each model. To assess which factors most influenced each of the three independent variables within each model, was used the R package “vip” (permutation-based VI scores method) developed in 2018 [ 44 ]. The individual conditional expectation (ICE) curves were subsequently constructed to underline the relationship between the most important variables in the models and the output of the model.…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Variables Influencing per Capita Production, Separate Collection, and Costs of Municipal Solid Waste in the Apulia Region (Italy): An Experience of Deep Learning

Fasano

Addante

Valenzano

et al. 2021

IJERPH

View full text Add to dashboard Cite

Municipal solid waste (MSW) must be managed to reduce its impact on environmental matrices and population health as much as possible. In particular, the variables that influence the production, separate waste collection, and costs of MSW must be understood. Although many studies have shown that such factors are specific to an area, the awareness of these factors has created opportunities to implement operations to enable more effective and efficient MSW management services, and to specifically respond to the variables that have the most impact. The deep learning approaches used in this study are effective in achieving this goal and can be used in any other territorial context to ensure that the organizations that deal with these issues are more aware and create useful plans to promote the circular economy. Our findings indicate the important influence of number of rooms in a residential buildings and construction years on MSW production, the combination of services such as municipal collection centers and door-to-door service for separate MSW collection and the characteristics of the residential buildings in the municipalities on MSW management costs.

show abstract

Section: Discussionsupporting

confidence: 93%

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Variables Influencing per Capita Production, Separate Collection, and Costs of Municipal Solid Waste in the Apulia Region (Italy): An Experience of Deep Learning

Fasano

Addante

Valenzano

et al. 2021

IJERPH

View full text Add to dashboard Cite

show abstract

“…Our first model-agnostic method is based on a simple feature importance ranking measure (FIRM); for details, see Greenwell et al (2018), Zien et al (2009), and Scholbeck et al (2019). The specific approach used here is based on quantifying the "flatness" of the effects of each feature.…”

Section: Variance-based Methodsmentioning

confidence: 99%

Variable Importance Plots—An Introduction to the vip Package

Greenwell¹,

Boehmke²

2020

The R Journal

268

161

View full text Add to dashboard Cite

In the era of "big data", it is becoming more of a challenge to not only build state-of-the-art predictive models, but also gain an understanding of what's really going on in the data. For example, it is often of interest to know which, if any, of the predictors in a fitted model are relatively influential on the predicted outcome. Some modern algorithms-like random forests (RFs) and gradient boosted decision trees (GBMs)-have a natural way of quantifying the importance or relative influence of each feature. Other algorithms-like naive Bayes classifiers and support vector machines-are not capable of doing so and model-agnostic approaches are generally used to measure each predictor's importance. Enter vip, an R package for constructing variable importance scores/plots for many types of supervised learning algorithms using model-specific and novel model-agnostic approaches. We'll also discuss a novel way to display both feature importance and feature effects together using sparklines, a very small line chart conveying the general shape or variation in some feature that can be directly embedded in text or tables. 1 Although "interpretability" is difficult to formally define in the context of ML, we follow Doshi-Velez and Kim (2017) and describe "interpretable" as the ".. . ability to explain or to present in understandable terms to a human." 2 In this context "importance" can be defined in a number of different ways. In general, we can describe it as the extent to which a feature has a "meaningful" impact on the predicted outcome. A more formal definition and treatment can be found in van der Laan (2006). 3 We refer the reader to Poulin et al. (2006), Caruana et al. (2015), Bibal and Frénay (2016), and Bau et al. (2017), for discussions around model structure interpretation.

show abstract

“…To evaluate the order of importance of the variables and related graphic processing, the permutation-based VI scores method [36] was used. For this method, in particular, the package "vip" version 0.1.3.9000 of the program R (Brandon Greenwell, Cincinnati, Ohio) was used [37].…”

Section: Statistical Analysesmentioning

confidence: 99%

Adherence to Vaccination Policy among Public Health Professionals: Results of a National Survey in Italy

et al. 2020

View full text Add to dashboard Cite

Starting from 2013, the number of unvaccinated people alarmingly increased in Italy; therefore, in 2017 a new Vaccine National Plan was approved. Healthcare workers (HCWs), especially public health professionals (PHPs, i.e., workers in in the sector of hygiene and preventive medicine), have an important role in informing and promoting vaccinations. In this context, the Italian Study Group of Hospital Hygiene of the Italian Society of Hygiene, Preventive Medicine and Public Health (GISIO-SItI) conducted a national survey to assess knowledge, attitude, and practices towards recommended vaccinations among PHPs. The survey was conducted during October 2019 with an anonymous questionnaire distributed to PHPs attending the 52° SItI National Congress. Overall, 57.1% of operators answered correctly to all seven recommended vaccinations, 12.8% reported to be vaccinated for all seven recommended vaccinations, while 30% were naturally immunized. A higher immunization coverage was reported for anti-hepatitis B (88.9%) and measles (86.1%), and 81.3% of the participants reported being offered the influenza vaccination during the 2018/2019 season. The majority of our sample indicated that hepatitis B (95%) and influenza (93.7%) were the recommended vaccines for HCWs, while less was known regarding varicella, pertussis, diphtheria, and tetanus boosters every 10 years. PHPs who were vaccinated (or who intended to be vaccinated) were more likely to recommend vaccinations to their patients and provided a reassuring example to those hesitant patients. Finally, this is the first study that identified good algorithms (using the techniques of machine learning as Random Forest and Deep Learning) to predict the knowledge of PHPs regarding recommended vaccinations with possible applications in other national and international contexts.

show abstract

Sampling, Intervention, Prediction, Aggregation: A Generalized Framework for Model-Agnostic Interpretations

Cited by 30 publications

References 18 publications

Variables Influencing per Capita Production, Separate Collection, and Costs of Municipal Solid Waste in the Apulia Region (Italy): An Experience of Deep Learning

Variables Influencing per Capita Production, Separate Collection, and Costs of Municipal Solid Waste in the Apulia Region (Italy): An Experience of Deep Learning

Variable Importance Plots—An Introduction to the vip Package

Adherence to Vaccination Policy among Public Health Professionals: Results of a National Survey in Italy

Contact Info

Product

Resources

About