Artificial intelligence extracts key insights from legal documents to predict intimate partner femicide.

Legal documents serve as valuable repositories of information pertaining to crimes, encompassing not only legal aspects but also relevant details about criminal behaviors. To date and the best of our knowledge, no studies in the field examine legal documents for crime understanding using an Artificial Intelligence (AI) approach. The present study aims to fill this research gap by identifying relevant information available in legal documents for crime prediction using Artificial Intelligence (AI). This innovative approach will be applied to the specific crime of Intimate Partner Femicide (IPF). A total of 491 legal documents related to lethal and non-lethal violence by male-to-female intimate partners were extracted from the Vlex legal database. The information included in these documents was analyzed using AI algorithms belonging to Bayesian, functions-based, instance-based, tree-based, and rule-based classifiers. The findings demonstrate that specific information from legal documents, such as past criminal behaviors, imposed sanctions, characteristics of violence severity and frequency, as well as the environment and situation in which this crime occurs, enable the correct detection of more than three-quarters of both lethal and non-lethal violence within male-to-female intimate partner relationships. The obtained knowledge is crucial for professionals who have access to legal documents, as it can help identify high-risk IPF cases and shape strategies for preventing crime. While this study focuses on IPF, this innovative approach has the potential to be extended to other types of crimes, making it applicable and beneficial in a broader context.

A Comprehensive Analysis of Factors Associated with Intimate Partner Femicide: A Systematic Review.

There has been a growing concern about violence against women by intimate partners due to its incidence and severity. This type of violence is a severe problem that has taken the lives of thousands of women worldwide and is expected to continue in the future. A limited amount of research exclusively considers factors related only to these women's deaths. Most focus on deaths of both men and women in an intimate partnership and do not provide precise results on the phenomenon under study. The necessity for an actual synthesis of factors linked solely to women's deaths in heterosexual relationships is key to a comprehensive knowledge of that case. This could assist in identifying high-risk cases by professionals involving an interdisciplinary approach. The study's objective is to systematically review the factors associated with these deaths. Twenty-four studies found inclusion criteria extracted from seven databases (Dialnet, Web of Science, Pubmed, Criminal Justice, Psychology and Behavioral Science Collection, Academic Search Ultimate, and APA Psyarticles). The review was carried out under the PRISMA guidelines' standards. The studies' quality assessment complies with the MMAT guidelines. Findings revealed that there are specific factors of the aggressor, victim, partner's relationship, and environment associated with women's deaths. The results have implications for predicting and preventing women's deaths, providing scientific knowledge applied to develop public action programs, guidelines, and reforms.

Risk Assessment Instruments for Intimate Partner Femicide: A Systematic Review.

Intimate partner violence is a severe problem that has taken the lives of thousands of women worldwide, and it is bound to continue in the future. Numerous risk assessment instruments have been developed to identify and intervene in high-risk cases. However, a synthesis of specific instruments for severe violence against women by male partners has not been identified. This type of violence has specific characteristics compared to other forms of intimate partner violence, requiring individualized attention. A systematic review of the literature has been conducted to summarize the intimate partner homicide risk assessment instruments applied to this population. It has been carried out with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement guidelines. The search strategy yielded a total of 1,156 studies, and only 33 studies met eligibility criteria and were included in the review. The data of these studies were extracted, analyzed, and presented on study characteristics (country and year, sample, data sources, purpose of the studies) and main findings (a brief description of the instruments, risk factor items, psychometric properties). The results indicate that the Danger Assessment, the Danger Assessment for Immigrants, the Danger Assessment for Law Enforcement, the Danger Assessment-5, the Taiwan Intimate Partner Violence Danger Assessment, the Severe Intimate Partner Risk Prediction Scale, The Lethality Screen, and the H-Scale are specific risk assessment instruments for predicting homicide and attempted homicide. There are differences in the number and content of risk assessment items, but most of them include the evidence's critical factors associated with homicide. Validity and reliability scores of these instruments vary, being consistency and accuracy medium-high for estimating homicide. Finally, implications for prediction and prevention are noted, and future research directions are discussed.

High-throughput micropatterning platform reveals Nodal-dependent bisection of peri-gastrulation-associated versus preneurulation-associated fate patterning.

In vitro models of postimplantation human development are valuable to the fields of regenerative medicine and developmental biology. Here, we report characterization of a robust in vitro platform that enabled high-content screening of multiple human pluripotent stem cell (hPSC) lines for their ability to undergo peri-gastrulation-like fate patterning upon bone morphogenetic protein 4 (BMP4) treatment of geometrically confined colonies and observed significant heterogeneity in their differentiation propensities along a gastrulation associable and neuralization associable axis. This cell line-associated heterogeneity was found to be attributable to endogenous Nodal expression, with up-regulation of Nodal correlated with expression of a gastrulation-associated gene profile, and Nodal down-regulation correlated with a preneurulation-associated gene profile expression. We harness this knowledge to establish a platform of preneurulation-like fate patterning in geometrically confined hPSC colonies in which fates arise because of a BMPs signalling gradient conveying positional information. Our work identifies a Nodal signalling-dependent switch in peri-gastrulation versus preneurulation-associated fate patterning in hPSC cells, provides a technology to robustly assay hPSC differentiation outcomes, and suggests conserved mechanisms of organized fate specification in differentiating epiblast and ectodermal tissues.

Global Sensitivity Estimates for Neural Network Classifiers.

Artificial neural networks (ANNs) have traditionally been seen as black-box models, because, although they are able to find ``hidden'' relations between inputs and outputs with a high approximation capacity, their structure seldom provides any insights on the structure of the functions being approximated. Several research papers have tried to debunk the black-box nature of ANNs, since it limits the potential use of ANNs in many research areas. This paper is framed in this context and proposes a methodology to determine the individual and collective effects of the input variables on the outputs for classification problems based on the ANOVA-functional decomposition. The method is applied after the training phase of the ANN and allows researchers to rank the input variables according to their importance in the variance of the ANN output. The computation of the sensitivity indices for product unit neural networks is straightforward as those indices can be calculated analytically by evaluating the integrals in the ANOVA decomposition. Unfortunately, the sensitivity indices associated with ANNs based on sigmoidal basis functions or radial basis functions cannot be calculated analytically. In this paper, the indices for those kinds of ANNs are proposed to be estimated by the (quasi-) Monte Carlo method.

Toward a study of gene regulatory constraints to morphological evolution of the Drosophila ocellar region.

The morphology and function of organs depend on coordinated changes in gene expression during development. These changes are controlled by transcription factors, signaling pathways, and their regulatory interactions, which are represented by gene regulatory networks (GRNs). Therefore, the structure of an organ GRN restricts the morphological and functional variations that the organ can experience-its potential morphospace. Therefore, two important questions arise when studying any GRN: what is the predicted available morphospace and what are the regulatory linkages that contribute the most to control morphological variation within this space. Here, we explore these questions by analyzing a small "three-node" GRN model that captures the Hh-driven regulatory interactions controlling a simple visual structure: the ocellar region of Drosophila. Analysis of the model predicts that random variation of model parameters results in a specific non-random distribution of morphological variants. Study of a limited sample of drosophilids and other dipterans finds a correspondence between the predicted phenotypic range and that found in nature. As an alternative to simulations, we apply Bayesian networks methods in order to identify the set of parameters with the largest contribution to morphological variation. Our results predict the potential morphological space of the ocellar complex and identify likely candidate processes to be responsible for ocellar morphological evolution using Bayesian networks. We further discuss the assumptions that the approach we have taken entails and their validity.

An assessment of the barriers to the consumers' uptake of genetically modified foods: a neural network analysis.

BACKGROUND: This paper studies which of the attitudinal, cognitive and socio-economic factors determine the willingness to purchase genetically modified (GM) food, enabling the forecasting of consumers' behaviour in Andalusia, southern Spain. This classification has been made by a standard multilayer perceptron neural network trained with extreme learning machine. Later, an ordered logistic regression was applied to determine whether the neural network can outperform this traditional econometric approach. RESULTS: The results show that the highest relative contributions lie in the variables related to perceived risks of GM food, while the perceived benefits have a lower influence. In addition, an innovative attitude towards food presents a strong link, as does the perception of food safety. The variables with the least relative contribution are subjective knowledge about GM food and the consumers' age. The neural network approach outperforms the correct classification percentage from the ordered logistic regression. CONCLUSION: The perceived risks must be considered as a critical factor. A strategy to improve the GM food acceptance is to develop a transparent and balanced information framework that makes the potential risk understandable by society, and make them aware of the risk assessments for GM food in the EU. For its success, it is essential to improve the trust in EU institutions and scientific regulatory authorities.