Leveraging transfer learning with deep learning for crime prediction.

Crime remains a crucial concern regarding ensuring a safe and secure environment for the public. Numerous efforts have been made to predict crime, emphasizing the importance of employing deep learning approaches for precise predictions. However, sufficient crime data and resources for training state-of-the-art deep learning-based crime prediction systems pose a challenge. To address this issue, this study adopts the transfer learning paradigm. Moreover, this study fine-tunes state-of-the-art statistical and deep learning methods, including Simple Moving Averages (SMA), Weighted Moving Averages (WMA), Exponential Moving Averages (EMA), Long Short Term Memory (LSTM), Bi-directional Long Short Term Memory (BiLSTMs), and Convolutional Neural Networks and Long Short Term Memory (CNN-LSTM) for crime prediction. Primarily, this study proposed a BiLSTM based transfer learning architecture due to its high accuracy in predicting weekly and monthly crime trends. The transfer learning paradigm leverages the fine-tuned BiLSTM model to transfer crime knowledge from one neighbourhood to another. The proposed method is evaluated on Chicago, New York, and Lahore crime datasets. Experimental results demonstrate the superiority of transfer learning with BiLSTM, achieving low error values and reduced execution time. These prediction results can significantly enhance the efficiency of law enforcement agencies in controlling and preventing crime.

The Hamas massacre of Oct 7, 2023, and its aftermath, medical crimes, and the Lancet commission report on medicine, Nazism, and the Holocaust.

BACKGROUND: The report of the Lancet Commission on medicine, Nazism, and the Holocaust, released in November 2023, calls for this history to be required for all health professions education, to foster morally courageous health professionals who speak up when necessary. MAIN BODY: The report was released a month after Hamas' October 7 invasion of Israel, with the accompanying massacre of over 1200 people, taking of civilian hostages, and gender-based violence. These acts constitute crimes against humanity including genocide. Post-October 7, war in Gaza resulted, with a legitimate objective of Israel defending itself within international law. The authors discuss an accompanying Statement to the report condemning Hamas crimes and denouncing the perpetrators' use of their own civilians as human shields, including in healthcare facilities, and with the Hamas attack unleashing immense and ongoing suffering in Israel and beyond. With some exceptions, the medical literature shows a marked absence of condemnation of Hamas atrocities and includes unsubstantiated criticisms of Israel's military. A significant surge in global antisemitism including on university campuses since October 7, 2023, has occurred; and health professionals, according to the Commission, have a special responsibility to fight antisemitism and discrimination of all kinds. In this context, the authors discuss the controversy and criticism regarding diversity, equity, and inclusion education programs ("DEI") including such programs failing to protect Jews on campuses, especially as the U.S. President Biden's "The U.S. National Strategy to Counter Antisemitism," released in May 2023, calls for the inclusion of issues of antisemitism and religious discrimination within all DEI education programs. The authors support an evidence-based approach to the Hamas massacre, its aftermath and its relevance to health professionals both within medicine and their global citizenship, including refuting the international community accusations and anti-Israel libel. CONCLUSIONS: The report of the Lancet Commission on medicine, Nazism, and the Holocaust has striking relevance to the Hamas massacre of October 7, 2023 and its aftermath. This is further conveyed in an accompanying Statement, that describes the report's implications for contemporary medicine, including: 1) provision of skills required to detect and prevent crimes against humanity and genocide; (2) care for victims of atrocities; (3) upholding the healing ethos central to the practice of medicine; and (4) fostering history-informed morally courageous health professionals who speak up when necessary.

Application of artificial intelligence to eyewitness identification.

Artificial intelligence is already all around us, and its usage will only increase. Knowing its capabilities is critical. A facial recognition system (FRS) is a tool for law enforcement during suspect searches and when presenting photos to eyewitnesses for identification. However, there are no comparisons between eyewitness and FRS accuracy using video, so it is unknown whether FRS face matches are more accurate than eyewitness memory when identifying a perpetrator. Ours is the first application of artificial intelligence to an eyewitness experience, using a comparative psychology approach. As a first step to test system accuracy relative to eyewitness accuracy, participants and an open-source FRS (FaceNet) attempted perpetrator identification/match from lineup photos (target-present, target-absent) after exposure to real crime videos with varied clarity and perpetrator race. FRS used video probe images of each perpetrator to achieve similarity ratings for each corresponding lineup member. Using receiver operating characteristic analysis to measure discriminability, FRS performance was superior to eyewitness performance, regardless of video clarity or perpetrator race. Video clarity impacted participant performance, with the unclear videos yielding lower performance than the clear videos. Using confidence-accuracy characteristic analysis to measure reliability (i.e., the likelihood the identified suspect is the actual perpetrator), when the FRS identified faces with the highest similarity values, they were accurate. The results suggest FaceNet, or similarly performing systems, may supplement eyewitness memory for suspect searches and subsequent lineup construction and knowing the system's strengths and weaknesses is critical.

Legal actors' and laypersons' utility judgments of eyewitness lineup procedures and outcomes.

OBJECTIVE: Recent attempts to model the relative performances of eyewitness lineup procedures necessarily include theoretical assumptions about the various costs/benefits, or utilities, of different identification outcomes. We collected data to incorporate empirically derived utilities into such modeling as well as data on various stakeholders' views of lineup procedures as tertiary objectives. HYPOTHESES: This research was exploratory; therefore, we did not have a priori hypotheses. METHOD: We surveyed judges' (n = 70), prosecutors' (n = 28), police officers' (n = 82), and laypersons' (n = 191) opinions about eyewitness identification procedures and the utilities of outcomes of eyewitness identification procedures. We incorporated the utility judgments into models comparing the desirability of various lineup reforms and compared policy preferences between our samples. RESULTS: All samples frequently mentioned estimator and system variables in open-ended evaluations of lineup procedures, but legal samples mentioned system variables more often than did laypersons. Reflector variables (e.g., confidence) were mentioned less often across the board, as was the scientific basis/standardization of identification policy (especially among laypersons). Utility judgments of various identification outcomes indicated that judges adopt values more closely aligned with normative legal ethics (i.e., the Blackstone ratio), whereas other stakeholders (especially laypersons) depart significantly from those standards. Utility models indicated general agreement among samples in lineup procedure preferences, which varied as a function of culprit-presence base rates. CONCLUSION: Although legal stakeholders vary in how they value eyewitness identification outcomes, their values imply relatively consistent policy preferences that sometimes depart from scientific recommendations. Nonetheless, all samples expressed support for using scientific research to inform legal policy regarding eyewitness evidence. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Do inconclusive forensic decisions disadvantage the innocent?

OBJECTIVE: Two experiments examined the potential for inconclusive forensic decisions to disadvantage the innocent. HYPOTHESES: Both experiments tested the hypothesis that inconclusive decisions produce more incriminating legal judgments than do clearly exculpatory forensic decisions. Experiment 2 also examined whether this hypothesized effect conformed to a confirmation bias, a communication error, or perceptual accuracy. METHOD: In Experiment 1 (N = 492), a forensic expert testified that physical evidence recovered from a crime scene either matched or did not match a suspect's evidence or produced an inconclusive result. In Experiment 2 (N = 1,002), a forensic expert testified that physical evidence recovered from a crime scene either matched or did not match a suspect's evidence, produced an inconclusive result, or was unsuitable for analysis. A fifth condition omitted the forensic evidence and expert testimony. RESULTS: The inconclusive decision produced less incriminating legal judgments than did the match forensic decision (|d|average = 0.96), more incriminating legal judgments than did the no-match forensic decision (|d|average = 0.62), and equivalent legal judgments to the unsuitable decision (|d|average = 0.12) and to legal judgments made in the absence of forensic evidence (|d|average = 0.07). These results suggest that participants interpreted the inconclusive decision to be forensically neutral, which is consistent with a communication error. CONCLUSION: The findings provide preliminary support for the idea that inconclusive decisions can put the innocent at risk of wrongful conviction by depriving them of a clearly exculpatory forensic decision. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Advancing the Shift-of-Strategy approach: Shifting suspects' strategies in extended interviews.

OBJECTIVE: Interviewers often face the challenge of obtaining information from suspects who are willing to speak but are motivated to conceal incriminating information. The Shift-of-Strategy (SoS) approach is an interviewing technique designed to obtain new information from such suspects. This study provides a robust empirical test of the SoS approach using more complex crime events and longer interviews than previously tested as well as testing a new variation of the approach (SoS-Reinforcement) that included a strategic summary of the suspect's statement. We compared this new variation with a standard version of the approach (SoS-Standard) and an interviewing approach that involved no confrontation of discrepancies in the suspects' statements (Direct). HYPOTHESES: We predicted that the two SoS versions would outperform the Direct condition in terms of participants' disclosure of previously unknown information. We also predicted that SoS-Reinforcement would outperform SoS-Standard. Finally, we expected that participants in the SoS conditions would not assess the interview or the interviewer more poorly than participants in Direct. METHOD: A total of 300 participants completed an online mock crime procedure, and they were subsequently interviewed with one of the three interviewing techniques. Following the interview, participants provided assessments of their experiences being interviewed. RESULTS: Participants in both SoS-Standard (d = 0.49, 95% confidence interval [CI: 0.21, 0.78]) and SoS-Reinforcement (d = 0.59, 95% CI [0.30, 0.87]) disclosed more previously unknown information than participants in the Direct condition, but SoS-Reinforcement did not outperform SoS-Standard (d = 0.08, 95% CI [-0.20, 0.36]). Participants in SoS-Reinforcement assessed their experience more negatively than those in Direct. No such differences were observed in the remaining two-way comparisons. CONCLUSIONS: The study provides support for the effectiveness of eliciting new information through the SoS approach and illuminates possible experiential downsides with being subjected to the SoS-Reinforcement approach. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Examining the effectiveness of the Gateway conditional caution on health and well-being of young adults committing low-level offences: a randomised controlled trial.

BACKGROUND: Young adults who commit low-level offences commonly have a range of health and social needs and are significantly over-represented in the criminal justice system. These young adults may need to attend court and potentially receive penalties including imprisonment. Alternative routes exist, which can help address the underlying causes of offending. Some feel more should be done to help young adults entering the criminal justice system. The Gateway programme was a type of out-of-court disposal developed by Hampshire Constabulary, which aimed to address the complex needs of young adults who commit low-level crimes. This study aimed to evaluate the effectiveness and cost-effectiveness of the Gateway programme, issued as a conditional caution, compared with usual process. METHODS: The Gateway study was a pragmatic, parallel-group, superiority randomised controlled trial that recruited young adults who had committed a low-level offence from four sites covering Hampshire and Isle of Wight. The primary outcome was mental health and well-being measured using the Warwick-Edinburgh Mental Well-being Scale. Secondary outcomes were quality of life, alcohol and drug use, and recidivism. Outcomes were measured at 4, 16 and 52 weeks postrandomisation. RESULTS: Due to issues with retention of participants and low data collection rates, recruitment ended early, with 191 eligible participants randomised (Gateway 109; usual process 82). The primary outcome was obtained for 93 (48.7%) participants at 4 weeks, 93 (48.7%) at 16 weeks and 43 (22.5%) at 1 year. The high attrition rates meant that effectiveness could not be assessed as planned. CONCLUSIONS: Gateway is the first trial in a UK police setting to have a health-related primary outcome requiring individual data collection, rather than focusing solely on recidivism. We demonstrated that it is possible to recruit and randomise from the study population, however follow-up rates were low. Further work is needed to identify ways to facilitate engagement between researchers and vulnerable populations to collect data. TRIAL REGISTRATION NUMBER: ISRCTN11888938.

Using risk of crime detection to study change in mechanisms of decision making.

Perceptions of crime detection risk (e.g., risk of arrest) play an integral role in the criminal decision-making process. Yet, the sources of variation in those perceptions are not well understood. Do individuals respond to changes in legal policy or is perception of detection risk shaped like other perceptions-by experience, heuristics, and with biases? We applied a developmental perspective to study self-reported perception of detection risk. We test four hypotheses against data from the Dunedin Longitudinal Study (analytic sample of N = 985 New Zealanders), a study that spans 20 years of development (Ages 18-38, years 1990-2011). We reach four conclusions: (1) people form their perception of detection risk early in the life course; (2) perception of detection risk may be general rather than unique to each crime type; (3) population-level perceptions are stable between adolescence and adulthood; but (4) people update their perceptions when their life circumstances change. The importance of these findings for future theoretical and policy work is considered. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Effectiveness of substance use disorder treatment as an alternative to imprisonment.

INTRODUCTION: Drug courts are criminal justice programs to divert people with substance use disorders from incarceration into treatment. Drug courts have become increasingly popular in the US and other countries. However, their effectiveness in reducing important public health outcomes such as recidivism and substance-related health harms remains ambiguous and contested. We used nationwide register data from Sweden to evaluate the effectiveness of contract treatment sanction, the Swedish version of drug court, in reducing substance misuse, adverse somatic and mental health outcomes, and recidivism. METHODS: In this prospective cohort study, two quasi-experimental designs were used: difference-in-differences and the within-individual design. In the latter, we compared the risk of outcomes during time on contract treatment to, 1) parole after imprisonment and, 2) probation. RESULTS: The cohort included 11,893 individuals (13% women) who underwent contract treatment. Contract treatment was associated with a reduction of 7 percentage points (95% CI: -.088, -.055) in substance misuse, 5 percentage points (-.064, -.034) in adverse mental health events, 9 percentage points (-.113, -.076) in adverse somatic health events, and 3 fewer charges (-3.16, -2.85) for crime in difference-in-differences analyses. Within-individual associations suggested that the same individual had longer times-to-event for all outcomes during contract treatment than on parole or on probation. CONCLUSIONS: Contract treatment is an effective intervention from both public health and criminal justice perspective. Our findings suggest that it is a superior alternative to incarceration in its target group. Further, we find that an implementation approach that is less punitive and more inclusive than what is typical in the US can be successful.

Analysing Crime Scene Data, Medical Records and Forensic Information to Determine the Causes of Suicides.

OBJECTIVE:  To analyse crime scene data, medical records, and forensic information to unveil insights into the causes and traits of suicides. STUDY DESIGN:  Descriptive study. Place and Duration of the Study: Department of Family Medicine, Kayseri City Hospital, Kayseri, Turkiye, between January 2020 to December 2021. METHODOLOGY:  A suicide investigation team (doctor, social worker, psychologist) was created to study cases and conduct on-site psychological autopsies. Triggered by emergency calls, the team interviewed suicide victims' relatives using semi-structured questionnaires, gathering data on personal details, time, method, and potential motives. Medical records revealed psychiatric history and medication use, while national judicial systems were reviewed for legal records. RESULTS:  A total of 158 fatal suicides were studied. Males accounted for 73.4%, females 26.6%. The leading cause was psychiatric illness (43%), chiefly depression (39%). Suicide peaked in the fall, especially in September, mainly at 23:00-23:59. Home was the common site (58.9%), and hanging was the primary method (44.3%). Prior hospitalisation for suicide attempts was 7.5%. Criminal records were held by 16.4% (26 individuals). CONCLUSION:  The results support the idea that suicides have seasonal patterns and that there are temporal windows of increased risk for suicide. KEY WORDS:  Suicide reasons, Suicide time, Psychological autopsy, Seasonal and temporal patterns.

A location discrete choice model of crime: Police elasticity and optimal deployment.

Despite the common belief that police presence reduces crime, there is mixed evidence of such causal effects in major Latin America cities. In this work we identify the casual relationship between police presence and criminal events by using a large dataset of a randomized controlled police intervention in Bogotá D.C., Colombia. We use an Instrumental Variables approach to identify the causal effect of interest. Then we consistently estimate a Conditional Logit discrete choice model with aggregate data that allow us to identify agents' utilities for crime location using Two Stage Least Squares. The estimated parameters allow us to compute the police own and cross-elasticities of crime for each of the spatial locations and to evaluate different police patrolling strategies. The elasticity of crime to police presence is, on average across spatial locations, -0.26 for violent crime, -0.38 for property crime and -0.38 for total crime, all statistically significant. Estimates of cross-elasticities are close to zero; however, spillover effects are non-negligible. Counterfactual analysis of different police deployment strategies show, for an optimal allocating algorithm, an average reduction in violent crime of 7.09%, a reduction in property crimes of 8.48% and a reduction in total crimes of 5.15% at no additional cost. These results show the potential efficiency gains of using the model to deploy police resources in the city without increasing the total police time required.

Lower autonomic arousal as a risk factor for criminal offending and unintentional injuries among female conscripts.

BACKGROUND: Lower autonomic arousal is a well-known correlate of criminal offending and other risk-taking behaviors in men, but few studies have investigated this association in women. AIM: To test associations between autonomic arousal and criminal offending as well as unintentional injuries among female conscripts. METHODS: All women born 1958-1994 in Sweden who participated in voluntary military conscription (n = 12,499) were identified by linking Swedish population-based registers. Predictors were resting heart rate (RHR) and systolic blood pressure (SBP). Covariates were height, weight, and physical energy capacity. Main outcomes were criminal convictions (any, violent, and non-violent) from the National Crime Register. Secondary outcome was unintentional injuries requiring medical treatment or causing death. We used survival analyses to test for associations between predictors and outcomes. RESULTS: Low RHR, relative to high RHR, was associated with an increased risk of any criminal conviction, non-violent criminal convictions, and unintentional injuries. Low SBP, relative to high SBP, was associated with an increased risk of violent criminal convictions. CONCLUSIONS: Results support lower autonomic arousal, particularly lower RHR, as a correlate of criminal offending among women that warrants further examination, as the reported findings have potential implications for the prediction of future female crime.

Cancer cell invasion: Caveolae and invadosomes are partners in crime.

During cancer progression, tumor cells need to disseminate by remodeling the extracellular tumor matrix. A recent study sheds light on the intricate cooperation between caveolae and invadosomes that facilitates the spread of cancer cells.

Examining the stability and change in age-crime relation in South Korea, 1980-2019: An age-period-cohort analysis.

The aggregate-level age-crime distributions in Western countries are predominantly right-skewed and adolescent-spiked. Based on Western data, Hirschi and Gottfredson (1983) asserted that this age-crime pattern is universally invariant across time and places. This study's overall goal is to rigorously examine Hirschi and Gottfredson's invariant premise within a non-Western country, focusing on the stability and change in the age-crime patterns of South Korea from 1980 to 2019. Specifically, two research questions are addressed: (1) whether the average age-arrest curves in South Korea diverge from the invariant premise after adjusting for period and cohort effects; (2) how period and cohort effects modify the age-arrest curves. To examine these questions, I applied the age-period-cohort-interaction model (APC-I) to analyze the official age-specific arrest statistics for various offense types from 1980 to 2019 in South Korea. Findings suggested that the age-crime patterns of homicide, assault, and fraud are characterized by spread-out distributions and advanced peak ages. After adjusting for period and cohort effects, most of the age-crime curves are still robustly divergent from the age-crime distributions found in Western countries. Cohort and period effects have modified the age-crime patterns, but arrests in South Korea are largely concentrated among midlife age groups older than 30. These results provide additional compelling evidence contesting Hirschi and Gottfredson's invariance thesis, underscoring the substantial impact of country-specific processes, historical context, and cultural factors on the age-crime relationship.

Assessing the impact and implications of the revised Act on the Aggravated Punishment of Specific Crimes in preventing child traffic injuries in school zones in Korea: an interrupted time series analysis.

In 2019, a child's death in Korea led to legislation that imposed stricter penalties for school zone traffic violations. We assessed the impact of that legislation using 2017-2022 Traffic Accident Analysis System data. Adjusted analyses revealed a significant decline in severe injuries in school zones, decreasing from 11 cases to 8 cases per month (p=0.017). The legislation correlated with a reduced risk of all child traffic injuries (risk ratio, 0.987; 95% confidence interval, 0.977 to 0.997; p=0.002), indicating its efficacy in curbing accidents.

Cross-border forensic profiling of fraudulent identity and travel documents: A pilot project between France and Switzerland.

The serial character of document fraud and its connection to organised crime groups who produce, sell and/or use fraudulent documents is a challenge for security and crime fighting. As a response, the added value of forensic intelligence is increasingly recognised. Using a forensic profiling method and a dedicated system deployed in Switzerland, document examiners can detect series (i.e., documents that share a common source) of fraudulent documents conveniently and efficiently. This detection can trigger or orientate investigations, supports crime intelligence efforts, and facilitates cross-jurisdictional cooperation. This study aims to assess the suitability of the forensic profiling system for international purpose and the efficiency of the method to detect cross-border series. The forensic profiling system has been deployed in France in the framework of a cross-border pilot project conducted by the School of Criminal Justice from the University of Lausanne and the French National Police (Division Nationale de Lutte contre la Fraude Documentaire et à l'Identité) over the period July 2019-May 2020. Data from the Swiss and French forensic profiling systems were compared to each other to detect cross-border series. The study sought to create operating conditions as close as possible to the real-life conditions of the profiling systems. The results are extremely positive both quantitatively and qualitatively. They demonstrate the benefit of setting up a systematic exchange of forensic data issued from profiling systems for fraudulent identity documents between France and Switzerland, let alone between any other countries. The results open up a very promising prospect for a sustained operational implementation by the police services of both countries and the extension of the exchanges internationally.