DAEDALUS at PAN 2014: Guessing tweet author's gender and age

This paper describes our participation at PAN 2014 author profiling task. Our idea was to define, develop and evaluate a simple machine learning classifier able to guess the gender and the age of a given user based on his/her texts, which could become part of the solution portfolio of the company. We were interested in finding not the best possible classifier that achieves the highest accuracy, but to find the optimum balance between performance and throughput using the most simple strategy and less dependent of external systems. Results show that our software using Naive Bayes Multinomial with a term vector model representation of the text is ranked quite well among the rest of participants in terms of accuracy.

Service for the pseudonymization of electronic healthcare records based on ISO/EN 13606 for the secondary use of information

The availability of electronic health data favors scientific advance through the creation of repositories for secondary use. Data anonymization is a mandatory step to comply with current legislation. A service for the pseudonymization of electronic healthcare record (EHR) extracts aimed at facilitating the exchange of clinical information for secondary use in compliance with legislation on data protection is presented. According to ISO/TS 25237, pseudonymization is a particular type of anonymization. This tool performs the anonymizations by maintaining three quasi-identifiers (gender, date of birth and place of residence) with a degree of specification selected by the user. The developed system is based on the ISO/EN 13606 norm using its characteristics specifically favorable for anonymization. The service is made up of two independent modules: the demographic server and the pseudonymizing module. The demographic server supports the permanent storage of the demographic entities and the management of the identifiers. The pseudonymizing module anonymizes the ISO/EN 13606 extracts. The pseudonymizing process consists of four phases: the storage of the demographic information included in the extract, the substitution of the identifiers, the elimination of the demographic information of the extract and the elimination of key data in free-text fields. The described pseudonymizing system was used in three Telemedicine research projects with satisfactory results. A problem was detected with the type of data in a demographic data field and a proposal for modification was prepared for the group in charge of the drawing up and revision of the ISO/EN 13606 norm.

Una mirada antropológica sobre el conflicto, la desigualdad y los procesos de reinserción social en Colombia

For more than half of a century, Colombia has been living in a state of violence, a nationwide political violence. As the time goes by, this situation gets even worse. Now the violence is implanted for different interests, such as personal, political, social and economical interests. For this reason, the information for this thesis was gathered from the time of the independence, through the era of violence until today; considering a reflection that begins with the perspective of Marx, passing by a theoretical compilation of the conflict, with an anthropological, psychological, biological and sociological perspective. In addition, different theories about violence have been studied to recognize the ideological approach of the different armed movements that have emerged in Colombia. Statistical, economical and social data including class struggle, social stratification, exclusion and gender perspective among others, have been considered from an anthropological and interdisciplinary approach for studying the violence. A strong interest in the social reintegration process, through an ethnographic study, based on the actors backgrounds, their lives, experiences and their geographical location have provided information that will allow to know the reality of those who are living in the process, those who survive it and those who go on with their lives in society, and those who after doing it return to the armed groups. Population displaced by violence, refugees, has contributed to this study about their possibilities to return to the society or to be excluded by it. With this purpose, a theoretical and a documentary analysis as well as fieldwork have been done, trying to bring forward tools and guidelines to make some progress in developing effective solutions for social integration of Colombian armed conflict victims.

Responses to Domestic Violence Public Service Ads: Memory, Attitudes, Affect, and Individual Differences

Public service ads (PSAs) are an increasingly visible part of efforts to decrease the occurrence and consequences of domestic violence. Like other advertising, domestic violence PSAs are designed to grab attention, influence attitudes, and enhance memory for ad content. Over the years, images in domestic violence PSAs have changed substantially; agencies have started using pictures that generate emotions - either vivid negative images (bruised faces or body parts), or positive images (smiling faces) that contrast with the negative text. It is not clear, however, how different types of ad images influence memory for the message and attitudes about domestic violence, and what role affect may play in such responses. Moreover, the extent to which individual differences (trauma history, posttraumatic distress - PTSD symptoms) influence outcomes is not known. In three studies with undergraduate and community samples, using methods ranging from psychophysiology to self-report, the impact of images on attitudes and memory for ad content are investigated, also considering affect and individual differences. Results indicate graphic negative images enhanced memory for ad content, are rated as more persuasive, and are more likely to compel the viewer to act. Affective responses to ads also differed based on image type, and in some cases, partially mediated the relationship between ads and outcomes. Trends in the data suggest further study of the role of individual differences (trauma history, PTSD symptoms) is needed. This research provides information specifically relevant to the design of domestic violence public service campaigns and broadly relevant to understanding the role of emotional responses and individual differences on outcomes associated with public service ads.

The Relationship Between Feminine Gender Norm Conformity and Purpose and Meaning in Life Among Female College Students.

This study explores the relationship between conformity to traditional feminine gender norms and meaning and purpose in life among female college students. Based on previous research findings that show a negative correlation between specific traditional feminine gender norm conformity factors and psychological well-being, we posited that participants with higher levels of traditional gender norm conformity will experience a lower sense of meaning and purpose in life. Self-report data was obtained from 338 female undergraduate college students using the Conformity to Feminine Norms Inventory (CFNI) and the Purpose in Life Test (PIL). A standard multiple regression assessed the accuracy of eight feminine norms in predicting purpose and meaning in life. Results partially supported the hypothesis, with Modesty as the most significant contributor to meaning and purpose. Inconsistent with our hypothesis, participants who endorsed high scores on Nice in Relationships, Involvement with Children, Sexual Fidelity, and Domestic, scored higher on the PIL.

Ethnic Violence on Kenya's Periphery: Informal Institutions and Local Resilience in Conflict-Affected Communities

Deadly, inter-ethnic group conflict remains a threat to international security in a world where the majority of armed violence occurs not only within states but in the most ungoverned areas within states. Conflicts that occur between groups living in largely ungoverned areas often become deeply protracted and are difficult to resolve when the state is weak and harsh environmental conditions place human security increasingly under threat. However, even under these conditions, why do some local conflicts between ethnic groups escalate, whereas others do not? To analyze this puzzle, the dissertation employs comparative methods to investigate the conditions under which violence erupts or stops and armed actors choose to preserve peace. The project draws upon qualitative data derived from semi-structured interviews, focus group dialogues, and participant observation of local peace processes during field research conducted in six conflict-affected counties in Northern Kenya. Comparative analysis of fifteen conflict episodes with variable outcomes reveals the conditions under which coalitions of civic associations, including local peace committees, faith-based organizations, and councils of elders, inter alia, enhance informal institutional arrangements that contain escalation. Violence is less likely to escalate in communities where cohesive coalitions provide platforms for threat-monitoring, informal pact making, and enforcement of traditional codes of restitution. However, key scope conditions affect whether or not informal organizational structures are capable of containing escalation. In particular, symbolic acts of violence and the use of indiscriminant force by police and military actors commonly undermine local efforts to contain conflict. The dissertation contributes to the literatures on civil society and peacebuilding, demonstrating the importance of comparing processes of escalation and non-escalation and accounting for interactive effects between modes of state and non-state response to local, inter-ethnic group conflict.

Social determinants and health effects of low and high severity intimate partner violence

PURPOSE: We sought to analyze whether the sociodemographic profile of battered women varies according to the level of severity of intimate partner violence (IPV), and to identify possible associations between IPV and different health problems taking into account the severity of these acts. METHODS: A cross-sectional study of 8,974 women (18-70 years) attending primary healthcare centers in Spain (2006-2007) was performed. A compound index was calculated based on frequency, types (physical, psychological, or both), and duration of IPV. Descriptive and multivariate procedures using logistic regression models were fitted. RESULTS: Women affected by low severity IPV and those affected by high severity IPV were found to have a similar sociodemographic profile. However, divorced women (odds ratio [OR], 8.1; 95% confidence interval [CI], 3.2-20.3), those without tangible support (OR, 6.6; 95% CI, 3.3-13.2), and retired women (OR, 2.7; 95% CI, 1.2-6.0) were more likely to report high severity IPV. Women experiencing high severity IPV were also more likely to suffer from poor health than were those who experienced low severity IPV. CONCLUSIONS: The distribution of low and high severity IPV seems to be influenced by the social characteristics of the women involved and may be an important indicator for estimating health effects. This evidence may contribute to the design of more effective interventions.

Gender-sensitive reporting in medical research

Sex and gender differences influence the health and wellbeing of men and women. Although studies have drawn attention to observed differences between women and men across diseases, remarkably little research has been pursued to systematically investigate these underlying sex differences. Women continue to be underrepresented in clinical trials, and even in studies in which both men and women participate, systematic analysis of data to identify potential sex-based differences is lacking. Standards for reporting of clinical trials have been established to ensure provision of complete, transparent and critical information. An important step in addressing the gender imbalance would be inclusion of a gender perspective in the next Consolidated Standards of Reporting Trials (CONSORT) guideline revision. Uniform Requirements for Manuscripts Submitted to Biomedical Journals, as a set of well-recognized and widely used guidelines for authors and biomedical journals, should similarly emphasize the ethical obligation of authors to present data analyzed by gender as a matter of routine. Journal editors are also promoters of ethical research and adequate standards of reporting, and requirements for inclusion of gender analyses should be integrated into editorial policies as a matter of urgency.

Gender analysis of moxifloxacin clinical trials

Purpose: To determine the inclusion of women and the sex-stratification of results in moxifloxacin Clinical Trials (CTs), and to establish whether these CTs considered issues that specifically affect women, such as pregnancy and use of hormonal therapies. Previous publications about women’s inclusion in CTs have not specifically studied therapeutic drugs. Although this type of drug is taken by men and women at a similar rate, adverse effects occur more frequently in the latter. Methods: We reviewed 158 published moxifloxacin trials on humans, retrieved from MedLine and the Cochrane Library (1998–2010), to determine whether they complied with the gender recommendations published by U.S. Food and Drug Administration Guideline. Results: Of a total of 80,417 subjects included in the moxifloxacin CTs, only 33.7% were women in phase I, in contrast to phase II, where women accounted for 45%, phase III, where they represented 38.3% and phase IV, where 51.3% were women. About 40.9% (n = 52) of trials were stratified by sex and 15.3% (n = 13) and 9% (n = 7) provided data by sex on efficacy and adverse effects, respectively. We found little information about the influence of issues that specifically affect women. Only 3 of the 59 journals that published the moxifloxacin CTs stated that authors should stratify their results by sex. Conclusions: Women are under-represented in the published moxifloxacin trials, and this trend is more marked in phase I, as they comprise a higher proportion in the other phases. Data by sex on efficacy and adverse effects are scarce in moxifloxacin trials. These facts, together with the lack of data on women-specific issues, suggest that the therapeutic drug moxifloxacin is only a partially evidence-based medicine.

Mapping and exploring health systems' response to intimate partner violence in Spain

Background: For a comprehensive health sector response to intimate partner violence (IPV), interventions should target individual and health facility levels, along with the broader health systems level which includes issues of governance, financing, planning, service delivery, monitoring and evaluation, and demand generation. This study aims to map and explore the integration of IPV response in the Spanish national health system. Methods: Information was collected on five key areas based on WHO recommendations: policy environment, protocols, training, monitoring and prevention. A systematic review of public documents was conducted to assess 39 indicators in each of Spain’s 17 regional health systems. In addition, we performed qualitative content analysis of 26 individual interviews with key informants responsible for coordinating the health sector response to IPV in Spain. Results: In 88% of the 17 autonomous regions, the laws concerning IPV included the health sector response, but the integration of IPV in regional health plans was just 41%. Despite the existence of a supportive national structure, responding to IPV still relies strongly on the will of health professionals. All seventeen regions had published comprehensive protocols to guide the health sector response to IPV, but participants recognized that responding to IPV was more complex than merely following the steps of a protocol. Published training plans existed in 43% of the regional health systems, but none had institutionalized IPV training in medical and nursing schools. Only 12% of regional health systems collected information on the quality of the IPV response, and there are many limitations to collecting information on IPV within health services, for example underreporting, fears about confidentiality, and underuse of data for monitoring purposes. Finally, preventive activities that were considered essential were not institutionalized anywhere. Conclusions: Within the Spanish health system, differences exist in terms of achievements both between regions and between the areas assessed. Progress towards integration of IPV has been notable at the level of policy, less outstanding regarding health service delivery, and very limited in terms of preventive actions.

Social determinants of sexual satisfaction in Spain considered from the gender perspective

Aims: To describe gender- and social class-related inequalities in sexual satisfaction and analyze their relationship with self-perceived health status. Methods: This population-based, cross-sectional study included 7384 sexually active people aged 16 years and over residing in Spain in 2009 (3951 men and 3433 women). The explanatory variables were gender, age, social class, share in performing domestic tasks, spend time looking after oneself, collaborate economically in supporting the family, caring for children, self-perceived health status, and the desire to increase or decrease frequency of having sexual relations. Bivariate and multivariate logistic regression models were fitted. Results: Among women, sexual satisfaction declines progressively after age 45. Sexual satisfaction is 1.7 times higher among women who look after themselves and who feel good compared with those who do not. The odds of wanting to increase sex is 3.3 times higher for women who are satisfied compared with women who desire a lower frequency of sexual intercourses; and good perceived health was associated with sexual satisfaction. In satisfied men, the corresponding odds is 1.9 times that of men desiring to reduce their frequency of sex. Conclusions: Gender and social class inequalities are found in sexual satisfaction. This is associated with perceived health status, adding evidence in support of the World Health Organization definition of sexual health.

Economic crisis, immigrant women and changing availability of intimate partner violence services: a qualitative study of professionals' perceptions in Spain

Introduction: Since 2008, Spain has been in the throes of an economic crisis. This recession particularly affects the living conditions of vulnerable populations, and has also led to a reversal in social policies and a reduction in resources. In this context, the aim of this study was to explore intimate partner violence (IPV) service providers’ perceptions of the impact of the current economic crisis on these resources in Spain and on their capacity to respond to immigrant women’s needs experiencing IPV. Methods: A qualitative study was performed based on 43 semi-structured in-depth interviews to social workers, psychologists, intercultural mediators, judges, lawyers, police officers and health professionals from different services dealing with IPV (both, public and NGO’s) and cities in Spain (Barcelona, Madrid, Valencia and Alicante) in 2011. Transcripts were imported into qualitative analysis software (Atlas.ti), and analysed using qualitative content analysis. Results: We identified four categories related to the perceived impact of the current economic crisis: a) “Immigrant women have it harder now”, b) “IPV and immigration resources are the first in line for cuts”, c) “ Fewer staff means a less effective service” and d) “Equality and IPV policies are no longer a government priority”. A cross-cutting theme emerged from these categories: immigrant women are triply affected; by IPV, by the crisis, and by structural violence. Conclusion: The professionals interviewed felt that present resources in Spain are insufficient to meet the needs of immigrant women, and that the situation might worsen in the future.

Factors Associated with the Cessation of Intimate Partner Violence in Women Attending Primary Care in Spain

This study aimed to identify factors associated with the likelihood of IPV cessation among women attending Spanish primary healthcare. Of the 2465 women who reported lifetime IPV, 36.1 % stated that violence had ceased. Those women not currently abused had higher levels of education and social support, were workers or students, and had no dependent children. When IPV duration was less than 5 years, the likelihood of cessation was two times higher than when IPV continued beyond 5 years. For women who have experienced physical IPV, the probability of ending the violent relationship was 10 times higher than for those suffering from psychological IPV. The implications of the findings regarding clinical significance and future research are discussed.

Applying the WHO recommendations on health-sector response to violence against women to assess the Spanish health system. A mixed methods approach

This methodological note describes the development and application of a mixed-methods protocol to assess the responsiveness of Spanish health systems to violence against women in Spain, based on the World Health Organization (WHO) recommendations. Five areas for exploration were identified based on the WHO recommendations: policy environment, protocols, training, accountability/monitoring, and prevention/promotion. Two data collection instruments were developed to assess the situation of 17 Spanish regional health systems (RHS) with respect to these areas: 1) a set of indicators to guide a systematic review of secondary sources, and 2) an interview guide to be used with 26 key informants at the regional and national levels. We found differences between RHSs in the five areas assessed. The progress of RHSs on the WHO recommendations was notable at the level of policies, moderate in terms of health service delivery, and very limited in terms of preventive actions. Using a mixed-methods approach was useful for triangulation and complementarity during instrument design, data collection and interpretation.

Infancia y violencia de género: cuantificación y percepciones

Cada vez más autores destacan la necesaria incorporación de la infancia en el estudio de la violencia de género y, en concreto, de la violencia contra la mujer por parte del compañero íntimo (VCI). Tal incorporación puede facilitar no sólo una menor invisibilidad y desprotección de niños y niñas, sino también una mejor comprensión de la propia violencia, sus características y dinámicas. Partiendo de estas consideraciones, en este artículo se presentan los resultados de un análisis secundario de los datos obtenidos en la última macroencuesta sobre violencia contra las mujeres en España. La información disponible permite analizar: 1) la presencia de menores expuestos a VCI; 2) la posible relación entre esta presencia y la formulación de denuncias; y 3) la percepción que tienen las mujeres víctimas de VCI en torno al rol parental de los agresores.