Domestic violence risk prediction in Iran using a machine learning approach by analyzing Persian textual content in social media

Domestic violence (DV) against women in Iran is a hidden societal issue. In addition to its chronic physical, mental, industrial, and economic effects on women, children, and families, DV prevents victims from receiving mental health care. On the other hand, DV campaigns on social media have encouraged victims and society to share their stories of abuse. As a result, massive amount of data has been generated about this violence, which can be used for analysis and early detection. Therefore, this study aimed to analyze and classify Persian textual content pertinent to DV against women in social media. It also aimed to use machine learning to predict the risk of this content. After collecting 53,105 tweets and captions in the Persian language from Twitter and Instagram, between April 2020 and April 2021, 1611 tweets and captions were chosen at random and categorized using criteria compiled and approved by an expert in the field of DV. Then, using machine learning algorithms, modeling and evaluation processes were performed on the tagged data. The Naïve Base model, with an accuracy of 86.77% was the most accurate model among all machine learning models for predicting critical Persian content pertinent to domestic violence on social media. The obtained findings indicate that using a machine learning approach, the risk of Persian content related to DV in social media against women can be predicted.

Application of complexity theory in health and social care research: a scoping review

BACKGROUND

Complexity theory has been chosen by many authors as a suitable lens through which to examine health and social care. Despite its potential value, many empirical investigations apply the theory in a tokenistic manner without engaging with its underlying concepts and underpinnings.

OBJECTIVES

The aim of this scoping review is to synthesise the literature on empirical studies that have centred on the application of complexity theory to understand health and social care provision.

METHODS

This scoping review considered primary research using complexity theory-informed approaches, published in English between 2012 and 2021. Cochrane Database of Systematic Reviews, MEDLINE, CINAHL, EMBASE, Web of Science, PSYCHINFO, the NHS Economic Evaluation Database, and the Health Economic Evaluations Database were searched. In addition, a manual search of the reference lists of relevant articles was conducted. Data extraction was conducted using Covidence software and a data extraction form was created to produce a descriptive summary of the results, addressing the objectives and research question. The review used the revised Arksey and O'Malley framework and adhered to the Preferred Reporting Items for Systematic Reviews and Meta-analysis Extension for Scoping Reviews (PRISMA-ScR).

RESULTS

2021 studies were initially identified with a total of 61 articles included for extraction. Complexity theory in health and social care research is poorly defined and described and was most commonly applied as a theoretical and analytical framework. The full breadth of the health and social care continuum was not represented in the identified articles, with the majority being healthcare focused.

DISCUSSION

Complexity theory is being increasingly embraced in health and care research. The heterogeneity of the literature regarding the application of complexity theory made synthesis challenging. However, this scoping review has synthesised the most recent evidence and contributes to translational systems research by providing guidance for future studies.

CONCLUSION

The study of complex health and care systems necessitates methods of interpreting dynamic prcesses which requires qualitative and longitudinal studies with abductive reasoning. The authors provide guidance on conducting complexity-informed primary research that seeks to promote rigor and transparency in the area.

REGISTRATION

The scoping review protocol was registered at Open Science Framework, and the review protocol was published at BMJ Open (https://bit.ly/3Ex1Inu).

Daily Reports of Aggressive Behaviors in Interpersonal Conflicts

Interpersonal conflicts are inevitable, but the probability that conflicts involve aggressive behavior varies. Prior research that has tended to focus on victimization in intimate partnerships reported through retrospective designs. Addressing these limitations, the current study examines daily reports of behaving aggressively in any conflict across relationships in a sample of 512 young adults drawn from the nationally representative iCOR cohort. Respondent attitudes and affective measures were collected at the end of the daily data collection period. Regression methods were applied to examine the probability and frequency of aggression, investigating early and recent exposure to adversities, attitudes, self-control, affect and emotional states, and alcohol use behavior. Recent adversities and the propensity to endorse a defensive honor code attitude, consistent with theory and retrospective studies of aggression, predicted both prevalence and frequency of aggressive behavior. The associations of childhood maltreatment and self-control with the prevalence of behaving aggressively were as expected, but these constructs were significantly associated with the frequency of aggression with unexpected, inverse directionality. Moreover, respondents' affect and other emotional states were only associated with the frequency, not the prevalence, of aggressive behavior. Overall, this daily data collection constructively distinguished risk and protective factors for behaving aggressively more often. Further research is needed to disentangle the extent to which affective states drive or is a consequence of frequent aggressive behavior.

Modeling Women's Need For Action in Violent Relationships

We are beginning to understand that intimate partner violence (IPV) and women's decision-making about that violence are nonlinear phenomena. IPV and decision-making are influenced by variables feedforwarding upon themselves with multiple interconnected predictors and circularly causal relationships. Computer models can help us gain a systems perspective on these relationships and enable hypothesis-testing without engendering risk to women in these relationships. The purpose of this study was to develop a mathematical model of women's decision-making concerning her violent relationship and assess the impact of random stress and her controllable behaviors on violence and decision-making. An agent-based model was created using data from couples with history of violence, based upon results of multiple time series of partner violence. To explore factors that may alter model results, eight continuous variable parameters were created based upon significant (p ≤ .05) but discrepant (opposite directions) results from two prior time series studies. Overall, 13 unique patterns of violence in five categories were identified, but none of these categories included his violence alone without some additional influence (i.e., marital distance leading to marital distance the following day). To assess the potential impact that random stress and behaviors under her control (arguments, forgiveness, alcohol use, violence) could have on need-for-action and actions taken, the effects of variable parameter settings on these outcomes were also assessed. While random stress had little effect on outcomes, her interventions could have an impact but were pattern-specific. Her daily participation in arguments correlated with more violence. The need-for and actually taking action were at times independent of each other. This mathematical model yielded results that generally involved her violence with or without his violence. Thus, modeling partner violence and women's decision-making is possible, yielding diverse patterns. However, the complexity of interdependent predictors unique to each relationship means that targeted interventions will need to be couple-specific.

Violence towards women and their decisions to take action: A complex systems approach

Intimate partner violence (IPV) is a blight on society. Our traditional understanding suggests that interventions should be straightforward, leading to predictable positive results. However, these assumptions do not reflect the reality of IPV, which continues to frustrate physicians. While IPV research has thoroughly described the developmental risks and impacts of IPV, the violent incident itself remains largely unstudied and poorly understood. Although this lack of research may partially explain physician frustration and the limitations of our interventions, the greater problem may be our reliance upon the wrong paradigm in guiding our understanding. Complexity science says that systems are globally understandable, but not completely knowable. Our hypothesis is that IPV needs to be viewed as a complex adaptive system if we are to understand the phenomenon, identify expectations and appropriately intervene. When viewed through the lens of complexity science, IPV becomes less knowable and predictable, suggesting that interventionists should expect variable response. Research has indeed demonstrated that partner violence is a complex phenomenon with multiple, interdependent factors and a nonlinear trajectory. This nonlinearity/unpredictability can impact outcomes in IPV, often more so than the frequency or severity of the violence. Similarly, women's decision-making concerning the violence is also a nonlinear process dependent upon multiple factors and catastrophic influences. Once recognized, complexity science offers a novel approach to explain IPV's obfuscation and resistance to predictable change. Using the tempered expectations of a systems lens, the violent interdependencies can be clarified, the obscure causes of events can be visualized, and the temporal irregularities can be mapped. Not only can the disruptive tipping points, system feedforward propagations, powerful attractors and discontinuities compromise reasoned intervention, but these same factors, if understood, can be harnessed to foster and magnify circumstances that enable positive change.

Is Perceived Need for Action Among Women in Violent Relationships Nonlinear and, If So, Why?

Despite the prevalence and impact of partner violence, we understand little about women's action taking except that it seems an unpredictable, nonlinear process. This article determines the degree of nonlinearity in perceived need for help, legal action, or leaving among women in violent relationships. The participants included 143 women who experienced violence in the previous month, enrolled from six primary care clinics. Baseline surveys assessed background characteristics and factors which may affect perceived need for action. Multiple times series assessments of violence and need for action were collected daily for 8 weeks via telephone Interactive Voice Response. Measures of nonlinearity of violence, perceived need for help, legal action, and leaving were computed. Repeated measures ANOVA assessed differences across measures of nonlinearity. To identify factors contributing to nonlinearity, staged multiple regression assessed the relationship between nonlinearity measures and outcomes. Ninety-three women completed sufficient time series for nonlinearity assessment. Measures of nonlinearity were lower for need for legal action compared with needs for help and leaving. Regression analysis suggested that isolation, social networks, and lack of awareness contribute to nonlinearity. Women's perceived need for legal action and its level of nonlinearity were lowest compared with those of help seeking and leaving. Although its relative linearity suggests that the need for legal action may be the most predictable, its lower mean rating suggests that legal action is a low priority. Although need for help and leaving are of higher priorities, their nonlinearity suggests that intervention will not yield predictable results.

Quantitative Methods for Analyzing Intimate Partner Violence in Microblogs: Observational Study

Background

Social media is a rich, virtually untapped source of data on the dynamics of intimate partner violence, one that is both global in scale and intimate in detail.

Objective

The aim of this study is to use machine learning and other computational methods to analyze social media data for the reasons victims give for staying in or leaving abusive relationships.

Methods

Human annotation, part-of-speech tagging, and machine learning predictive models, including support vector machines, were used on a Twitter data set of 8767 #WhyIStayed and #WhyILeft tweets each.

Results

Our methods explored whether we can analyze micronarratives that include details about victims, abusers, and other stakeholders, the actions that constitute abuse, and how the stakeholders respond.

Conclusions

Our findings are consistent across various machine learning methods, which correspond to observations in the clinical literature, and affirm the relevance of natural language processing and machine learning for exploring issues of societal importance in social media.

Is Readiness to Take Action Among Women in Violent Relationships a Catastrophic Phenomenon?

Taking action among women in violent relationships appears to involve sudden changes and reversals after periods of building stress, suggesting that decision making is a "catastrophic" phenomenon. This study sought to determine whether readiness-to-change is best modeled as a cusp catastrophic (CCM) phenomenon among women in violent relationships. A total of 143 women who experienced violence in the previous month completed baseline and end-of-study interviews assessing her hope, coping strategies, social network, and readiness-for-action (seeking help, taking legal action, and leaving) concerning the violence. Daily assessments of his violent behavior, forgiveness sought and given, and her perceived need-for-action were collected via telephone Interactive Voice Response for 8 weeks. Using regression analysis, the impact of factor-analyzed asymmetry (violence burden) and bifurcation (hope and cope, support, forgiveness, and number of children) variables on the outcomes (readiness-for-help, legal action, and leaving) was modeled, comparing the CCM against linear models to determine which model accounts for the most variance in each outcome. Cusp catastrophe models for all three actions accounted for more variance than either linear model comparison, but violence burden was only relevant to readiness-for-help and different bifurcation variables were at work for each action. While forgiveness was an important bifurcation factor in readiness-for-help and number of children served as the bifurcation factor for readiness-for-legal-action, readiness-to-leave was more complex with both number of children and hope-and-cope as bifurcation factors. Not only should we expect sudden changes in readiness but efforts to facilitate decision making should focus on addressing the bifurcation factors that may distort her interpretation of reality.

Predictors of Perceived Need for and Actual Action Taking Among Women in Violent Relationships

Decision-making of women in violent relationships is poorly understood. The study seeks to identify predictors of need-for-action and actions taken by women in violent relationships. The participants were 143 women who experienced violence in previous month from 6 primary care clinics. The methods involved multiple times series using daily assessments of household environment, marital relationship, concerns, violence, and need-for-action collected via telephone interactive voice response for 8 weeks. Outcomes include daily need-for-action and reports of actions taken. Same-day correlates and prior-day associations using vector autoregressions were sought, combined across subjects using meta-analytic techniques. Need for help depended on stalking, concern for child safety, forgiveness, and low perceived control; actually seeking help depended on sense of control with same-day stress and need for help. Need for legal action depended on concern for child safety and finances with desire to keep family together; actually taking legal action, correlated only with prior-day stalking and concerns about child safety but less about effects of violence on child. Need to leave depended on his violence, with concern about its effect on child, her forgiveness, and a low desire to keep family together, while actually leaving was primed by a day of his drinking, and triggered by same-day stress and need to leave, but lower levels of her drinking or his seeking forgiveness. Once gone, prior-day stalking and his alcohol use correlated with returning to the relationship. Taking action depends upon few prior- and same-day factors unique to each action.

Effects of religious and spiritual variables on outcomes in violent relationships

OBJECTIVE

Religious and spiritual factors in intimate partner violence have received increasing attention. But are such factors related to outcomes in violent relationships? The purpose of this study was to assess the relative impact of spiritual symptoms and religious coping on attitudinal/behavioral and clinical outcomes among women in violent relationships.

METHODS

Adult women with a recent history of husband-to-wife physical abuse were recruited from six primary care clinics. Once enrolled, 200 subjects completed a baseline interview and daily assessment of level of violence, using the Interactive Verbal Response for 12 weeks. At the completion of the study, contact with each participant was attempted to determine whether she had either sought professional help or left the relationship. Three religious/spiritual variables were assessed at baseline-number of visits to a religious/spiritual counselor, religious coping, and severity of spiritual symptoms. Stepped multiple linear regression was used to explain factor-analyzed outcomes (coping and appraisals, hope and support, symptomatology, functional status, readiness for change, and medical utilization), adjusting for demographic, marital, childhood, mental health, and violence variables.

RESULTS

After controlling for duration, severity and dynamics of violence, the use of spiritual resources, and the level of spiritual symptoms were associated with most attitudinal/behavioral and clinical outcomes, while religious coping was only associated with staying in the relationship.

CONCLUSIONS

Religious and spiritual factors were associated with most outcomes. Spiritual symptoms had a consistently negative effect on outcomes while use of spiritual resources had variable effects. Religious coping was only associated with refraining from leaving the relationship.

Assessment of an interactive voice response system for identifying falls in a statewide sample of older adults

OBJECTIVE

Interactive voice response (IVR) systems offer great advantages for data collection in large, geographically dispersed samples involving frequent contact. We assessed the quality of IVR data collected from older respondents participating in a statewide falls prevention program evaluation in Pennsylvania in 2010-12.

METHOD

Participants (n=1834) were followed up monthly for up to 10 months to compare respondents who completed all, some, or no assessments in the IVR system. Validity was assessed by examining IVR-reported falls incidence relative to baseline in-person self-report and performance assessment of balance.

RESULTS

While a third of the sample switched from IVR to in-person calls over follow-up, IVR interviews were successfully used to complete 68.1% of completed monthly assessments (10,511/15,430). Switching to in-person interviews was not associated with measures of participant function or cognition. Both self-reported (p<.0001) and performance assessment of balance (p=.05) at baseline were related to falls incidence.

CONCLUSION

IVR is a productive modality for falls research among older adults. Future research should establish what level of initial personal research contact is optimal for boosting IVR completion rates and what research domains are most appropriate for this kind of contact.

Multi-day recurrences of intimate partner violence and alcohol intake across dynamic patterns of violence

RATIONALE, AIMS AND OBJECTIVES

Consistent links exist between male alcohol intake and male-perpetrated intimate partner violence (IPV) as well as female alcohol intake and female-perpetrated IPV. However, the nature of the relationship remains unclear. This study attempted to identify unique alcohol-violence patterns within three different types of relationship dynamics to better understand the alcohol-violence relationship and its role in violence dynamics.

METHOD

Two hundred women in abusive relationships were recruited from six primary care clinics. Subjects completed daily assessments of their relationship using interactive verbal response via telephone for 12 weeks. Dynamic patterns (periodic, chaotic, random) were determined by positive versus negative Lyapunov exponents and measures of correlation dimension saturation. To identify recurrent day-to-day activities, we used orbital decomposition (based on symbolic dynamics).

RESULTS

Periodic dynamics included daily reports with mutual abuse and alcohol intake while random dynamics included a variety of patterns, especially those involving unequal mutual abuse. Unique strings for each dynamic pattern were examined. Periodic dynamics involved heavy alcohol intake by the husband or mutual moderate-severe violence. Random dynamics uniquely involved mutual verbal abuse with husband's alcohol intake on same or different days as well as husband-perpetrated moderate-severe violence with or without husband-perpetrated minor violence. Chaotic dynamics uniquely involved combinations from wife-perpetrated minor violence alone to combinations of husband's heavy alcohol intake (with or without husband-perpetrated minor violence), mutual verbal abuse, and husband-perpetrated verbal abuse (with or without husband's heavy alcohol intake).

CONCLUSION

Recurrent 4-day patterns were observed. Each dynamic pattern was characterized by recurrent strings unique to that pattern.

Webs of causation in violent relationships

RATIONALE, AIMS AND OBJECTIVES

Predictors of intimate partner violence (IPV) include husband, wife and relational characteristics. However, we know little about the proximal, day-to-day triggering of abusive events. The purpose of the study was to analyse the daily temporal relationships among environmental, relational and violence variables in violent marital relationships.

METHOD

Two hundred adult primary care women who experienced violence in the previous month were recruited from six primary care clinics. Women completed daily assessments of household environment, marital relationship and violence using telephone interactive verbal response for 12 weeks. Same-day correlates were sought using cross-correlations among the environmental, relational and violence factors. Prior-day and prior-week associations were sought using vector autoregressions.

RESULTS

Except for wife's alcohol intake, all household environment and relationship factors demonstrated significant same-day correlations with IPV. However, prior-day violence by the husband, hassles, lack of husband's alcohol intake, emotional upset and marital distance were significantly related to current husband-perpetrated violence. Wife's violence depended upon her prior-day violence and alcohol intake only. All factors were related to husband-perpetrated violence in the subsequent week. Only wife's alcohol intake and husband's seeking forgiveness did not feedforward. In addition to the presence of multiple interdependent factors, circular causality was noted for marital distance and feeling upset.

CONCLUSIONS

IPV was due to multiple interdependent factors, feedforward dynamics and circular causality as expected in complex systems. The complex dynamics imply that simple interventions may have little chance of success, but understanding couple-specific dynamics may allow women to recognize high-risk prior-day profiles and take preventive action.

Do violence dynamics matter?

RATIONALE, AIMS AND OBJECTIVES

Intimate partner violence is a complex, non-linear phenomenon. The purpose of this study was to determine whether violence dynamics (pattern, degree of non-linearity, optimal non-linearity) contributed to outcomes in violent relationships.

METHODS

The study was conducted in six primary care clinics, enrolling 200 adult women in violent relationships. In addition to baseline and end-of-study interviews, women completed daily telephone assessments of household environment and partner violence using interactive verbal response. Three non-linearity measures of violence were computed with 'optimal' non-linearity estimated using Z-transformations. Assignment of dynamic patterns (periodic, chaotic, random) was made based upon Lyapunov exponent and correlation dimension. Outcomes across dynamic patterns were analysed using analysis of variance. In addition, stepped multiple linear regression explained factor-analysed outcomes, adjusting for demographic, childhood, mental health and marital variables; attitudinal/behavioural outcomes were also adjusted for when explaining clinical outcomes.

RESULTS

Women experiencing periodic violence recognized the importance of violence and used their active coping to seek mental health care. Those with chaotic dynamics recognized that they were not responsible, experienced fewer psychological symptoms and emotional role limitations, and did not seek help. Those experiencing random violence recognized its unpredictability and uncontrollability. Violence non-linearity predicted negative coping, positive appraisals and hope/support in regression analyses, while optimal non-linearity contributed to readiness for change and symptoms functioning. Of the nine outcomes investigated, violence non-linearity contributed to five outcomes.

CONCLUSION

Dynamic pattern of violence, degree of violence non-linearity and optimal non-linearity correlated with several attitudinal/behavioural and clinical outcomes. Knowledge of violence dynamics may have applications when working with violent couples.