Benchmarking Biologically-Inspired Automatic Machine Learning for Economic Tasks

Lazebnik, Teddy; Fleischer, Torsten; Yaniv-Rosenfeld, Amit

doi:10.3390/su151411232

Cited by 7 publications

(3 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use the Tree-Based Pipeline Optimization Tool (TPOT), the genetic algorithm-based automatic machine learning library 45 . TPOT produces a full machine learning (ML) pipeline, including feature selection engineering, model selection, model ensemble, and hyperparameter tuning; and shown to produce impressive results in a wide range of applications [46][47][48] . Hence, for every configuration of source and target variables investigated, we used TPOT, allowing it to test up to 10000 ML pipelines.…”

Section: Classification and Regression Machine Learning Modelsmentioning

confidence: 99%

Digitally-Enhanced Dog Behavioral Testing: Getting Help from the Machine

Farhat

Lazebnik

Monteny

et al. 2023

Preprint

View full text Add to dashboard Cite

The assessment of behavioral traits in dogs is a well-studied challenge due to its many practical applications such as selection for breeding, prediction of working aptitude, chances of being adopted, etc. Most methods for assessing behavioral traits are questionnaire or observation-based, which require a significant amount of time, effort and expertise. In addition, these methods are also susceptible to subjectivity and bias, making them less reliable. In this study, we proposed an automated computational approach that may provide a more objective, robust and resource-efficient alternative to current solutions. Using part of a Stranger Test protocol, we tested (n=53) dogs for their response to the presence and benign actions of a stranger. Dog coping styles were scored by three experts. Moreover, data were collected from their handlers using the Canine Behavioral Assessment and Research Questionnaire (C-BARQ) . An unsupervised clustering of the dogs' trajectories revealed two main clusters showing a significant difference in the stranger-directed fear C-BARQ factor, as well as a good separation between (sufficiently) relaxed dogs and dogs with excessive behaviors towards strangers based on expert scoring. Based on the clustering, we obtained a machine learning classifier for expert scoring of coping styles towards strangers, which reached an accuracy of 78%. We also obtained a regression model predicting C-BARQ factor scores with varying performance, the best being Owner-Directed Aggression (with a mean average error of 0.108) and Excitability (with a mean square error of 0.032). This case study demonstrates a novel paradigm of digitally enhanced canine behavioral testing.

show abstract

Section: Classification and Regression Machine Learning Modelsmentioning

confidence: 99%

Digitally-Enhanced Dog Behavioral Testing: Getting Help from the Machine

Farhat

Lazebnik

Monteny

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Formally, given a dataset D∈R r,c with c∈N features and r∈N samples, we utilized TPOT, which uses a GA-based approach, to generate and test ML pipelines based on the popular scikit-learn library (38). Formally, we run the TPOT classifier search method to obtain an ML pipeline that aims to optimize the classifier's mean accuracy over the k-folds (39). Once the pipeline was obtained, we further aimed to improve the model's performance over the training cohort using the grid-search hyperparameters method (40) such that the hyperparameter value ranges were chosen manually (41).…”

Section: Detection Machine Learning Modelmentioning

confidence: 99%

Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data

Magana,

Gavojdian,

Menahem

et al. 2023

Front. Vet. Sci.

Self Cite

View full text Add to dashboard Cite

The present study aimed to employ machine learning algorithms based on sensor behavior data for (1) early-onset detection of digital dermatitis (DD) and (2) DD prediction in dairy cows. Our machine learning model, which was based on the Tree-Based Pipeline Optimization Tool (TPOT) automatic machine learning method, for DD detection on day 0 of the appearance of the clinical signs has reached an accuracy of 79% on the test set, while the model for the prediction of DD 2 days prior to the appearance of the first clinical signs, which was a combination of K-means and TPOT, has reached an accuracy of 64%. The proposed machine learning models have the potential to help achieve a real-time automated tool for monitoring and diagnosing DD in lactating dairy cows based on sensor data in conventional dairy barn environments. Our results suggest that alterations in behavioral patterns can be used as inputs in an early warning system for herd management in order to detect variances in the health and wellbeing of individual cows.

show abstract

“…For the method's performance validation, we manually collected 100 numerical datasets from Data World (https://data.world/datasets/economics; accessed on 24 October 2023) and Kaggle (https://www.kaggle.com; accessed on 24 October 2023), following [42]. The datasets were randomly chosen from a broad range of fields and represented a wide range of computational tasks.…”

Section: Performance Evaluationmentioning

confidence: 99%

Can We Mathematically Spot the Possible Manipulation of Results in Research Manuscripts Using Benford’s Law?

Lazebnik,

Gorlitsky

2023

Data

Self Cite

View full text Add to dashboard Cite

The reproducibility of academic research has long been a persistent issue, contradicting one of the fundamental principles of science. Recently, there has been an increasing number of false claims found in academic manuscripts, casting doubt on the validity of reported results. In this paper, we utilize an adapted version of Benford’s law, a statistical phenomenon that describes the distribution of leading digits in naturally occurring datasets, to identify the potential manipulation of results in research manuscripts, solely using the aggregated data presented in those manuscripts rather than the commonly unavailable raw datasets. Our methodology applies the principles of Benford’s law to commonly employed analyses in academic manuscripts, thus reducing the need for the raw data itself. To validate our approach, we employed 100 open-source datasets and successfully predicted 79% of them accurately using our rules. Moreover, we tested the proposed method on known retracted manuscripts, showing that around half (48.6%) can be detected using the proposed method. Additionally, we analyzed 100 manuscripts published in the last two years across ten prominent economic journals, with 10 manuscripts randomly sampled from each journal. Our analysis predicted a 3% occurrence of results manipulation with a 96% confidence level. Our findings show that Benford’s law adapted for aggregated data, can be an initial tool for identifying data manipulation; however, it is not a silver bullet, requiring further investigation for each flagged manuscript due to the relatively low prediction accuracy.

show abstract

Benchmarking Biologically-Inspired Automatic Machine Learning for Economic Tasks

Cited by 7 publications

References 79 publications

Digitally-Enhanced Dog Behavioral Testing: Getting Help from the Machine

Digitally-Enhanced Dog Behavioral Testing: Getting Help from the Machine

Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data

Can We Mathematically Spot the Possible Manipulation of Results in Research Manuscripts Using Benford’s Law?

Contact Info

Product

Resources

About