Weak Results Research Articles

Reliability of large language models as a tool for knowledge extraction from biographical dictionaries: the case of the Polish Biographical Dictionary

Abstract Large language models are tools with great potential for text processing. This study aims to assess the reliability of the models’ results in extracting structured knowledge from unstructured textual sources, particularly biographies from the Polish Biographical Dictionary. The task of the model was to extract information about the individuals, such as date and place of birth, death and burial, family relationships, important people, related settlements and institutions as well as occupied positions. The test was conducted on a sample of 250 biographies. The texts were written in Polish from the 1930s onwards and described the lives of individuals from various historical periods. The results show that the large language model (LLM) is very effective in identifying basic personal data, important family relationships, occupations, or offices held by the characters. Weaker results were obtained when attempting to find institutions and places associated with the protagonists. The outcome of the test suggests that LLMs can efficiently assist in digitizing and structuring historical biographical data and offer a promising tool for improving historical knowledge bases and speeding up the work compared to manual extraction of information.

Electrochemical Oxidation of Selected Micropollutants from Environment Matrices Using Boron-Doped Diamond Electrodes: Process Efficiency and Transformation Product Detection

Bisphenol A (BPA) and diclofenac (DCF) are among the most prevalent micropollutants in aquatic environments, with concentrations reaching up to several hundred µg/L. These compounds pose significant risks to biodiversity and environmental health, necessitating the development of effective removal methods. However, both BPA and DCF can be resistant to conventional treatment technologies, highlighting the need for innovative approaches. Electrochemical oxidation (EO) has emerged as a promising solution. In this study, we assessed the effectiveness of EO using boron-doped diamond (BDD) anodes to remove BPA and DCF from two types of treated wastewater (TWW-W and TWW-D) and landfill leachate (LL). The evaluation included an analysis of the removal efficiency of BPA and DCF and the identification of transformation products generated during the process. Additionally, the feasibility of the EO-BDD process to remove ammonium nitrogen (N-NH4+) and organic compounds present in these environmental matrices was investigated. The EO-BDD treatment achieved remarkable removal efficiencies, reducing BPA and DCF concentrations by over 96% in LL and TWW-W. Transformation product analyses identified four intermediates formed from parent compounds during the oxidation process. Furthermore, the EO-BDD process effectively removed both chemical oxygen demand (COD) and ammonium nitrogen from LL, although weaker results were observed for TWWs. These findings underscore the potential of the EO-BDD process as an effective method for the removal of BPA and DCF from challenging matrices, such as wastewater containing micropollutants. It also shows promise as a complementary technology for enhancing current conventional wastewater treatment methods, especially biological degradation.

Abstract PS10-03: Impact of Baseline Oestradiol and Testosterone on the Preventive Effect of Anastrozole

Abstract It is well known that serum levels of oestradiol and testosterone, esp free hormone levels, influence the risk of developing breast cancer in postmenopausal women (Thomas et al 1997, Hankinson et al 1998, Kaaks et al 2005, Tin Tin et al 2021). However very little is known about how these hormone levels influence the effectiveness of aromatase inhibitors. In the IBIS-II Prevention Trial we compared anastrozole to placebo in 3864 women at high risk of breast cancer (Cuzick et al 2020). Of these women 3644 (94.3%) had a baseline blood sample. In those with a valid blood sample, 72 in the anastrozole arm and 142 in the placebo arm developed breast cancer (including DCIS) after 12.9 years of follow up (OR = 0.49, 95% CI 0.37–0.66 P< 0.0001). For each case two controls were selected, matched on age, treatment arm and follow up longer than the matching case. In these women oestradiol (E2), testosterone (Testo) and SHBG were measured by liquid chromatography – tandem mass spectroscopy, and E2/SHBG and Testo/SHBG ratios were computed to approximate free hormone levels, and analysed in quartiles. Hormone replacement therapy was not allowed during the trial, and women with use within 3 months prior to entry or outlier hormone values were excluded from these analyses. In the placebo arm higher levels of both of these ratios were associated with a higher breast cancer rate (OR per quartile 1.25 (1.04-1.59), P=0.018) for E2/SHBG and (OR per quartile 1.22 (1.02-1.47), P = 0.032) for Testo/SHBG), whereas no significant effect was seen in the anastrozole arm (OR = 1.05 (0.81-1.37), and 1.16 (0.90-1.49), resp) (Table). Neither treatment interaction was significant. Absolute numbers of cases were similar between the anastrozole and placebo arms in the lowest quartile of E2/SHBG (18 anastrozole, 22 placebo), but higher numbers were seen in the placebo arm for the other quartiles. Similar results were seen for testo/SHBG, and supportive, but weaker results were seen for these two hormones without an SHBG adjustment, SHBG and BMI at entry. Adjusting for other risk factors had no influence on these findings. Effect sizes were similar during the 5 year treatment period and thereafter. There were too few ER negative cases to compare results by receptor status. Vasomotor side effects were higher with increasing E2/SHBG levels in both arms, but no other hormone showed an effect, and no hormone level was significantly associated with gynaecologic or musculoskeletal side effects. As aromatase inhibitors effectively eliminate the production of oestradiol in postmenopausal women, these results suggest they may be more effective in postmenopausal women with high oestradiol or testosterone levels, and raise questions about their value in women in the lower quartile of serum levels for these hormones for prevention and possibly adjuvant treatment. Table: Risk of breast cancer by quartiles of the oestradiol/SHBG in ratio high risk women treated with anastrozole or placebo. Citation Format: Jack Cuzick, Kim Chu, Brian Keevil, Antony Howell, Bernardo Bonanni, Evans Gareth, Kaija Holli, Sibylle Loibl, Nicholas Zdenkowski, Steven Cummings, Mitch Dowsett. Impact of Baseline Oestradiol and Testosterone on the Preventive Effect of Anastrozole [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS10-03.

An Experimental Analysis of Three-Phase Low-Voltage Power Factor Controllers Used in a Deforming Regime

In industry, to improve the power factor in low-voltage power substations, a power factor controller (the most used method) can be installed which connects capacitors banks (connected, or not, with coils) in the electrical installation. The most important parameters of power engineering are the power factors that indicate the efficiency of energy use. Currently, many non-linear consumers (more single-phase than three-phase) are used on low voltages. Harmonics (currents and/or voltages) are the most important dynamic component of power quality, affecting electrical equipment performance and also reducing power factors. The purpose of this analysis is to increase the displacement power factor and to decrease the total harmonic distortion (for the current) in the conditions where there are linear and non-linear consumers, where the power factor must be improved with capacitors banks. Relevant different consumers have been selected for both the industry and the home sector, as follows: inductive motors that are linear, inductive consumers, compact fluorescent lamps that are non-linear, and capacitive consumers. This analysis was carried out depending on the number of steps used for the power factor controller, the values of the capacitors banks, the AC reactor (connected in series with all consumers), and the LC shunt filters. For a slight deformation regime, a large number of capacitors banks with different values ensure a finer adjustment of the displacement factor. The maximum number of steps that regulators can command should not necessarily be used (the reliability of the installation decreases with the increase in the number of capacitors banks), but a reduced number of steps can be used, which can lead to higher values of displacement power factors. To improve the deforming regime and to increase the displacement power factor, the use of LC shunt filters, connected to a small number of steps, will also increase the displacement power factor (over 0.9) and decrease the total harmonic distortion (up to 7–10%) for the current. Weaker results were obtained with AC reactors connected to the power supply phases of consumers and, if a larger number of stages were used, to which LC shunt filters were connected, these filters become difficult to calibrate (resonances occur).

Soft Optimization of Treatment Methods for a Group of Neurological Diseases Using Data Mining

The growing healthcare industry is generating large amounts of useful data on patient demographics, treatment plans, payment, and insurance coverage, attracting the attention of clinicians and scientists alike. The purpose of this research is to predict diseases of the brain and nerves using data mining techniques. In this research, after data preparation, disease prediction has been attempted using large matrix methods and data mining techniques. By examining the new vector, we can find out which of the diseases in the matrix will be closer to this new disease with new symptoms using the rows of the matrix. The conducted research is one of descriptive-analytical and applied studies. In this research, we used different meters such as Manhattan, cosine similarity, Pearson, Minkowski and K nearest neighbor and implemented a program to predict neurological diseases using Python software. In the algorithm implemented by Python software, the doctor enters the symptoms of the patient and the program output of each meter shows three diseases close to the input symptoms and finally all the meters are compared and each time the meter is executed, which has a weaker result is determined. The advantages of each of these meters are explained below.

The relationship between game success indicators in the FIFA World Cup – Qatar 2022: An elastic-net regularization approach

Background and Study Aim. The significance of analyzing statistical indicators is paramount, as they provide valuable insights into the distinguishing performance factors of teams. The purpose of this study: 1) to identify statistically significant differences between relevant performance factors among winning, drawing, and losing teams in the 2022 Football World Cup in Qatar; 2) explore indicators that are most strongly associated with the game results in this tournament. Material and Methods. A total of 64 matches and 32 participating teams were analyzed. The variables included in the research were: total shots, shots on target, shots off target, effectiveness, passes, passes completed, crosses, crosses against, offsides committed, offsides received, fouls received, ball possession, fouls committed, corners, corners against, yellow cards and red cards. The results are extracted from the official FIFA website. The Analysis of Variance (ANOVA) and the ordinal logistic regression with elastic net were used for data calculation. Results. The results indicate that winning teams have statistically significantly higher mean values of variables related to goals scored: total shots (p < 0.05), shots on target (p < 0.01), and effectiveness (p < 0.01). According to the information-based model selection criteria AIC (Akaike information criterion) and BIC (Bayesian information criterion) effectiveness and shots on target are the most important variables regarding game results. They, along with corners, are best associated with successful teams. In contrast, crosses, ball possession, corners against, and yellow cards are associated with less successful teams. Results suggest that teams with high ball possession and large number of crosses, but fewer shots on target, have weaker results than teams with more shots on target and less ball possession. Conclusions. The study underscores the significance of certain performance indicators, such as total shots, shots on target, and effectiveness, in predicting a team's success. These findings provide valuable insights for coaches and teams to focus on specific aspects of their game to enhance their chances of winning. Additionally, understanding the impact of variables like ball possession, crosses, corners, and yellow cards emphasizes the importance of a well-rounded strategy for achieving success in international football tournaments.

Neighborhood Deprivation and Severe Maternal Morbidity in a Medicaid Population

Neighborhood Deprivation and Severe Maternal Morbidity in a Medicaid Population

Influence of Diet on the Bioavailability of Active Components from Zingiber officinale Using an In Vitro Digestion Model

Ginger (Zingiber officinale Rosc.) is a plant known all over the world that is used as a spice and as an ingredient in drinks, dietary supplements, and cosmetics. The growing availability of its fresh rhizomes makes it even more likely to be used in the diet, mainly due to its beneficial health properties and high content of polyphenols (gingerols and shogaols). The main goal and motivation of the authors was to assess the bioavailability of active substances contained in the extract from ginger rhizomes in the presence of various types of diets using the in vitro digestion method, enabling simulation of the processes occurring during the digestion and absorption of metabolites in the small intestine. For the qualitative and quantitative analyses, the HPLC-MS (High Performance Liquid Chromatography–Mass Spectrometry) and HPLC (High Performance Liquid Chromatography) techniques were used, respectively. Based on the obtained results, it was found that the best bioavailability of the selected ginger polyphenols (6-gingerol, 8-gingerdione, 8-shogaol, and 10-gingerdione) was estimated for a high-fiber diet, while the weakest results were obtained for standard and basic diets. In the case of the high-fiber diet, the bioavailability of the mentioned compounds was estimated as 33.3, 21.4, 6.73, and 21.0%, while for the basic diet, it was only 21.3, 5.3, 2.0, and 1.0%, respectively.

Influence of Solvents and Solvent Mixtures on the Content of Polyphenols and Antioxidant Activity of Capsella bursa-pastoris (L.) Medik. Extracts

Capsella bursa-pastoris (L.) Medik. (known as shepherd's purse) is a plant whose parts are used as medicine in herbal medicine. It is applicable as a medicine in the treatment of all forms of internal bleeding, for the treatment of hemorrhoids, excessive menstruation, but also for the usual stopping of nosebleeds. Through this research, the influence of organic solvents and their aqueous mixtures on the efficiency of polyphenol extraction and antioxidant activity was compared. The inhibition of free radicals was tested by the DPPH method, while the FRAP method was used to test the reduction potential. Analyzes have shown that water is the most effective solvent in the isolation of polyphenols from the aerial parts of shepherd's purse. Mixtures of organic solvents with water also showed high efficiency in the extraction of bioactive components, while the weakest results were obtained for extracts prepared in pure organic solvents.

Comparative applicability of MCDM‐SWOT based techniques for developing integrated watershed management framework

AbstractEnvironmental planning facilitates decision‐making to achieve sustainable development goals and provides a crucial way to achieve integrated watershed management (IWM). However, such systematic planning has not been adequately conducted worldwide. Therefore, this study was conducted to develop an IWM framework using SWOT (i.e., strengths, weaknesses, opportunities, and threats) model for the Cheshmeh–Kileh Watershed, Mazandaran Province, Iran. The input components were comparatively weighted using different multicriteria decision‐making (MCDM) techniques, including Game Theory Algorithm (GTA), Best–Worst Method (BWM), VIekriterijumsko KOmpromisno Rangiranje (VIKOR), Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), and Simple Additive Weighting (SAW). Semistructured interviews with stakeholders, including watershed residents, executive experts, and policymakers, were used to identify the SWOT factors. The number of interviewees in subgroups of local users, policy‐making institutions and executive organization were 75, 13, and 6 respectively. Five different MCDM techniques were then used to calculate the overall weight of SWOT factors. According to the results, the groups of SWOT factors of abundant water resources, severe floods, promotion of environmental conservation culture, and increasing migration of young age groups with respective weights of 0.298, 0.298, 0.372, and 0.279 scored higher than other factors. Based on the overall weights, it can be said that positive factors scored more points than negative factors. Therefore, according to the opinions of the stakeholders, the Cheshmeh–Kileh Watershed was in a good condition based on strengths and opportunities compared to weaknesses and threats, and for this reason, GTA and SAW, which had included the watershed status in the growth and development strategy, were selected as the best method. BWM and TOPSIS provided relatively acceptable results, and the weakest result was associated with VIKOR, which showed the watershed status in the defensive strategy. The present study results can help managers in optimal decisions for planning and optimal management in the Cheshmeh–Kileh Watershed to create ecological balance and increase the welfare of watershed residents.

Lying to Police Officers in the Ultimatum Game: An Experimental Study With Israeli Samples.

Two studies examined lying, fair sharing, and trust of Israeli police officers and laypeople to police and non-police target persons in the Ultimatum Game. Participants aimed to retain as many resources as possible in a sharing situation. To this end, they could conceal resources from the target person. Thus, a measure of lying was created by letting participants act in a specific role. Results indicated that police officers lied less to police targets than non-police targets. Conversely, laypeople lied more to police targets and less to non-police targets. Police officers' preference for honest sharing with police targets and laypeople's selfish sharing with police targets signified the first study's results. Results were explained by ingroup-outgroup differences, further inspired by severe events undermining the Israeli police's reputation. One year later, similar but weaker results were obtained in a second study. Police officers trusted police targets more than non-police targets, and laypeople trusted police targets less than non-police targets.

A Comparative Study of Motivation in Sports Training among Senior and Junior Professional Football Players in Bosnia and Herzegovina during the COVID-19 Pandemic

The Covid-19 pandemic has brought significant changes to the lives of athletes, particularly regarding the uncertainty surrounding the continuation of sports training and the postponement of sporting events. Given these disruptions to athletes’ routines, the role of motivation in sports participation is considered even more crucial. Therefore, this study aimed to examine the differences in motivation for sports participation during the Covid-19 pandemic between senior and junior professional football players. The sample consisted of 358 elite football players from the Premier League of Bosnia and Herzegovina. The players were divided into two groups: senior (n=193, 24.87±3.78 years) and junior (n=165, 18.08±1.00 years) football players. The Participation Motivation Questionnaire (PMQ) was utilized, comprising 19 items classified into six motivational dimensions: “Sports success”, “Social status”, “Friendship”, “Physical health”, “Sports activity”, and “Support”. A five-point Likert scale was used in the questionnaire, which was administered electronically via Google Forms. Results revealed that junior football players achieved significantly higher values in only three out of the 19 items: “Want to be popular” (.005), “Want to get status” (.001), and “Wished for friendly games” (.000). However, there were no differences between the senior and junior football player groups in the remaining 16 items. Also, both groups displayed exceptionally high motivation levels in the dimensions of “Sports success”, “Physical health”, and “’Friendship”, while the weakest results were observed in the dimension of ‘Support’ for both groups. Despite the specific challenges posed by the Covid-19 pandemic, the motivation to play football remains very high among both senior and junior football players. This study is essential as it provides insights into football players’ motivation and identifies differences in motivation levels between junior and senior players. The findings can be valuable for practical purposes, aiming to enhance motivation among football players, especially during circumstances where training conditions are disrupted, such as the Covid-19 pandemic.

Bonding Additively Manufactured PLA Materials: Effects of Joint Scarf Angle and Substrate Raster Orientation

The aim of this study is to investigate the effect of the scarf angle of the bonding region and the raster orientation of 3D printed substrates for the adhesive scarf joints made of additively manufactured Polylactic acid (PLA) adherends. In the first step, single PLA specimens were 3D printed using the fused filament fabrication (FFF) process in three different raster orientations of 0º, 45º, and 90º. Tensile and compression tests of them showed that the raster orientation of 90º gives the weakest results. Hence, in the second step, for the scarf joints, the PLA adherends were 3D printed in two different raster orientations of 0º and 45º. The joints were built with five different scarf angles of 15º, 30º, 45º, 60º, and 90º. The tensile and compression tests of all the specimens were conducted to determine the failure loads for different scarf angles and raster orientations. It is found that the endured load before rupture varies measurably as a function of the raster orientation of printed substrates and scarf angle of the joint, but differently for tensile and compression loadings. Eventually, the scarf angle for PLA scarf joints is determined which gives similar failure loads under both tension and compression loadings.

Combining interpolated maximum wind gust speed and forest vulnerability for rapid post-storm mapping of potential forest damage areas in Finland

Abstract In Finland, wind-induced forest damage is expected to increase in the future. Demand exists for timely and precise first-hand information about the main impact area of windstorms. Locating potential damage areas quickly is essential for effective operational planning of salvage loggings, aiming to reduce monetary losses of timber and risk for secondary damage caused by insects. This study presents an approach for mapping the potential damage areas immediately after a windstorm, by using high-resolution forest vulnerability data and information on the spatial distribution of maximum wind gust speed derived from weather station observations using kriging with external drift interpolation. The new method is evaluated by analyzing damage caused by nine major windstorms of the 2010s in Finland. Our results show that including roughness length information as an auxiliary variable in the interpolation improved the results and produced wind maps with more plausible structure and better separation between forested and non-forested land areas. The forest vulnerability data were most strongly linked to damage, whilst wind gust speed had weaker results. However, for future storms with unknown damage areas, we consider maximum wind gust speed still essential for defining the main impact area, whereas forest vulnerability data could then be used for more detailed damage predictions. Further advancements of wind interpolation approaches, preferably towards higher resolution and, if possible, based on a denser and more diverse observation network, is needed to fully exploit the potential of combined wind and forest vulnerability data. Albeit we recognize multiple uncertainties, room for improvements and benefits that additional data sources would bring, our study demonstrates a simple approach for rapid mapping of potential forest wind damage areas, which could be further developed into an operational tool.

Does a Nearby Murder Affect Housing Prices and Rents? The Case of Sydney

Does a Nearby Murder Affect Housing Prices and Rents? The Case of Sydney

Boosting ensembles for estimation of discharge coefficient and through flow discharge in broad-crested gabion weirs

Gabion weirs are environment-friendly structures widely used for irrigation and drainage network purposes. These structures' hydraulic performance is fundamentally different from solid weirs' due to their porosity and the existence of a through-flow discharge. This paper investigates the reliability and suitability of a number of Machine learning models for estimation of hydraulic performance of gabion weirs. Generally, three different Boosting ensemble models, including Gradient Boosting, XGBoost, and CatBoost, are compared to the well-known Random Forest and a Stacked Regression model, with respect to their accuracy in prediction of the discharge coefficient and through-flow discharge ratio of gabion weirs in free flow conditions. The Bayesian optimization approach is used to fine-tune model hyper-parameters automatically. Recursive feature elimination analysis is also performed to find optimum combination of features for each model. Results indicate that the CatBoost model has outperformed other models in terms of estimating the through flow discharge ratio (Qin/Qt) with R2 = 0.982, while both XGBoost and CatBoost models have shown close performance in terms of estimating the discharge coefficient (Cd) with R2 of CatBoost equal to 0.994 and R2 of XGBoost equal to 0.992. Weakest results were also produced by Decision tree regressor with R2 = 0.821 and 0.865 for estimation of Cd and Qin/Qt values.

A Dosiomics Analysis Based on Linear Energy Transfer and Biological Dose Maps to Predict Local Recurrence in Sacral Chordomas after Carbon-Ion Radiotherapy

Carbon Ion Radiotherapy (CIRT) is one of the most promising therapeutic options to reduce Local Recurrence (LR) in Sacral Chordomas (SC). The aim of this work is to compare the performances of survival models fed with dosiomics features and conventional DVH metrics extracted from relative biological effectiveness (RBE)-weighted dose (DRBE) and dose-averaged Linear Energy Transfer (LETd) maps, towards the identification of possible prognostic factors for LR in SC patients treated with CIRT. This retrospective study included 50 patients affected by SC with a focus on patients that presented a relapse in a high-dose region. Survival models were built to predict both LR and High-Dose Local Recurrencies (HD-LR). The models were evaluated through Harrell Concordance Index (C-index) and patients were stratified into high/low-risk groups. Local Recurrence-free Kaplan-Meier curves were estimated and evaluated through log-rank tests. The model with highest performance (median(interquartile-range) C-index of 0.86 (0.22)) was built on features extracted from LETd maps, with DRBE models showing promising but weaker results (C-index of 0.83 (0.21), 0.80 (0.21)). Although the study should be extended to a wider patient population, LETd maps show potential as a prognostic factor for SC HD-LR in CIRT, and dosiomics appears to be the most promising approach against more conventional methods (e.g., DVH-based).

A descriptive analysis of the characteristics, treatment response and prognosis of hepatic dominant solid tumors undergoing selective internal radiation therapy (SIRT)

Selective internal radiotherapy is widely used for liver dominant diseases of solid tumors. However, data about sequential treatment and prognostic factors are lacking. We consecutively included all 209 patients who received a selective internal radiotherapy intervention between January 2015 and May 2019. A retrospective analysis of their electronic patient records was performed regarding diagnosis of cancer, previous therapies and applied radioactive activity. A multicenter follow-up at least 6 weeks after intervention to assess radiological response and irregular subsequent follow-ups to asses disease progression were conducted. In addition, subgroup analyses were carried out. The most frequently treated indications were hepatocellular carcinoma (37%), colorectal cancers (14%), neuroendocrine tumors (9%), and breast cancer (8%). In hepatocellular carcinoma, selective internal radiotherapy was most performed without prior systemic therapy (40%), and for the remaining indications, most often after surgery with systemic therapy in sequence. Local radiological response, defined as either regression or stable disease, was assessed at least 6 weeks after intervention and showed 52% across all indications. Hepatocellular carcinoma (59%) and breast cancer (67%) showed an excellent, colorectal cancers (29%) a particularly poor response rate. Neuroendocrine tumors showed the third longest median post-selective internal radiation therapy (SIRT) survival with 12.4 months and the second longest median progression-free time with 5.2 months. Hepatocellular carcinoma showed even better results with a post-SIRT survival of 15.7 months and a median progression-free time of 5.3 months. Pancreatic neuroendocrine tumors showed significantly worse outcomes than other neuroendocrine tumors, regarding median post-SIRT survival and median progression-free time. No relevant SIRT related differences among sexes were detected. Patients with neuroendocrine tumors, breast cancer in late therapy lines and early-stage hepatocellular carcinoma seem to show better responses to SIRT than other entities. Colorectal cancers were mainly treated with SIRT in a second or third therapy line but with considerably weaker results than other entities.

Pride and Prejudice: Same-Sex Marriage Legalization Announcements and Hate Crimes

In this paper, we examine whether same-sex marriage legalization announcements impact the occurrence of hate crimes against lesbian, gay, bisexual, and transgender (LGBT) people. Using a difference-in-differences design, we exploit the variation in the timing of same-sex marriage legalization announcements across states. On average, a same-sex marriage legalization announcement reduces the anti-LGBT hate crime rate by .112 per 100,000 people, although some additional analyses have weaker results. Decreases are mostly driven by reductions in violent crimes. Event-study estimates show that results fade after 1 year and provide evidence that reductions are not due to changes in social trends before an announcement. Our results demonstrate that salient, progressive LGBT policy announcements may, by themselves, effectively reduce anti-LGBT hate crimes.

Density estimates for the zeros of the Beurling ζ function in the critical strip

We prove three results on the density resp. local density and clustering of zeros of the Beurling zeta function $\zeta(s)$ close to the one-line $\sigma:=\Re s=1$. The analysis here brings about some news, sometimes even for the classical case of the Riemann zeta function. Theorem 4 provides a zero density estimate, which is a complement to known results for the Selberg class. Note that density results for the Selberg class rely on use of the functional equation of $\zeta$, which we do not assume in the Beurling context. In Theorem 5 we deduce a variant of a well-known theorem of Tur\'an, extending its range of validity even for rectangles of height only $h=2$. In Theorem 6 we will extend a zero clustering result of Ramachandra from the Riemann zeta case. A weaker result -- which, on the other hand, is a strong sharpening of the average result from the classic book \cite{Mont} of Montgomery -- was worked out by Diamond, Montgomery and Vorhauer. Here we show that the obscure technicalities of the Ramachandra paper (like a polynomial with coefficients like $10^8$) can be gotten rid of, providing a more transparent proof of the validity of this clustering phenomenon.