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.
Integrating IoT devices in SCADA systems has provided efficient and improved data collection and transmission technologies. This enhancement comes with significant security challenges, exposing traditionally isolated systems to the public internet. Effective and highly reliable security devices, such as intrusion detection system (IDSs) and intrusion prevention systems (IPS), are critical. Countless studies used deep learning algorithms to design an efficient IDS; however, the fundamental issue of imbalanced datasets was not fully addressed. In our research, we examined the impact of data imbalance on developing an effective SCADA-based IDS. To investigate the impact of various data balancing techniques, we chose two unbalanced datasets, the Morris power dataset, and CICIDS2017 dataset, including random sampling, one-sided selection (OSS), near-miss, SMOTE, and ADASYN. For binary classification, convolutional neural networks were coupled with long short-term memory (CNN-LSTM). The system's effectiveness was determined by the confusion matrix, which includes evaluation metrics, such as accuracy, precision, detection rate, and F1-score. Four experiments on the two datasets demonstrate the impact of the data imbalance. This research aims to help security researchers in understanding imbalanced datasets and their impact on DL SCADA-IDS.
The proliferation of fake news has severe effects on society and individuals on multiple fronts. With fast-paced online content generation, has come the challenging problem of fake news content. Consequently, automated systems to make a timely judgment of fake news have become the need of the hour. The performance of such systems heavily relies on feature engineering and requires an appropriate feature set to increase performance and robustness. In this context, this study employs two methods for reducing the number of feature dimensions including Chi-square and principal component analysis (PCA). These methods are employed with a hybrid neural network architecture of convolutional neural network (CNN) and long short-term memory (LSTM) model called FakeNET. The use of PCA and Chi-square aims at utilizing appropriate feature vectors for better performance and lower computational complexity. A multi-class dataset is used comprising 'agree', 'disagree', 'discuss', and 'unrelated' classes obtained from the Fake News Challenges (FNC) website. Further contextual features for identifying bogus news are obtained through PCA and Chi-Square, which are given nonlinear characteristics. The purpose of this study is to locate the article's perspective concerning the headline. The proposed approach yields gains of 0.04 in accuracy and 0.20 in the F1 score, respectively. As per the experimental results, PCA achieves a higher accuracy of 0.978 than both Chi-square and state-of-the-art approaches.
Given the excessive foul language identified in audio and video files and the detrimental consequences to an individual's character and behaviour, content censorship is crucial to filter profanities from young viewers with higher exposure to uncensored content. Although manual detection and censorship were implemented, the methods proved tedious. Inevitably, misidentifications involving foul language owing to human weariness and the low performance in human visual systems concerning long screening time occurred. As such, this paper proposed an intelligent system for foul language censorship through a mechanized and strong detection method using advanced deep Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) through Long Short-Term Memory (LSTM) cells. Data on foul language were collected, annotated, augmented, and analysed for the development and evaluation of both CNN and RNN configurations. Hence, the results indicated the feasibility of the suggested systems by reporting a high volume of curse word identifications with only 2.53% to 5.92% of False Negative Rate (FNR). The proposed system outperformed state-of-the-art pre-trained neural networks on the novel foul language dataset and proved to reduce the computational cost with minimal trainable parameters.
Road traffic accidents have been listed in the top 10 global causes of death for many decades. Traditional measures such as education and legislation have contributed to limited improvements in terms of reducing accidents due to people driving in undesirable statuses, such as when suffering from stress or drowsiness. Attention is drawn to predicting drivers' future status so that precautions can be taken in advance as effective preventative measures. Common prediction algorithms include recurrent neural networks (RNNs), gated recurrent units (GRUs), and long short-term memory (LSTM) networks. To benefit from the advantages of each algorithm, nondominated sorting genetic algorithm-III (NSGA-III) can be applied to merge the three algorithms. This is named NSGA-III-optimized RNN-GRU-LSTM. An analysis can be made to compare the proposed prediction algorithm with the individual RNN, GRU, and LSTM algorithms. Our proposed model improves the overall accuracy by 11.2-13.6% and 10.2-12.2% in driver stress prediction and driver drowsiness prediction, respectively. Likewise, it improves the overall accuracy by 6.9-12.7% and 6.9-8.9%, respectively, compared with boosting learning with multiple RNNs, multiple GRUs, and multiple LSTMs algorithms. Compared with existing works, this proposal offers to enhance performance by taking some key factors into account-namely, using a real-world driving dataset, a greater sample size, hybrid algorithms, and cross-validation. Future research directions have been suggested for further exploration and performance enhancement.
Human action recognition (HAR) is one of the most active research topics in the field of computer vision. Even though this area is well-researched, HAR algorithms such as 3D Convolution Neural Networks (CNN), Two-stream Networks, and CNN-LSTM (Long Short-Term Memory) suffer from highly complex models. These algorithms involve a huge number of weights adjustments during the training phase, and as a consequence, require high-end configuration machines for real-time HAR applications. Therefore, this paper presents an extraneous frame scrapping technique that employs 2D skeleton features with a Fine-KNN classifier-based HAR system to overcome the dimensionality problems.To illustrate the efficacy of our proposed method, two contemporary datasets i.e., Multi-Camera Action Dataset (MCAD) and INRIA Xmas Motion Acquisition Sequences (IXMAS) dataset was used in experiment. We used the OpenPose technique to extract the 2D information, The proposed method was compared with CNN-LSTM, and other State of the art methods. Results obtained confirm the potential of our technique. The proposed OpenPose-FineKNN with Extraneous Frame Scrapping Technique achieved an accuracy of 89.75% on MCAD dataset and 90.97% on IXMAS dataset better than existing technique.
Brain tumor detection at early stages can increase the chances of the patient's recovery after treatment. In the last decade, we have noticed a substantial development in the medical imaging technologies, and they are now becoming an integral part in the diagnosis and treatment processes. In this study, we generalize the concept of entropy difference defined in terms of Marsaglia formula (usually used to describe two different figures, statues, etc.) by using the quantum calculus. Then we employ the result to extend the local binary patterns (LBP) to get the quantum entropy LBP (QELBP). The proposed study consists of two approaches of features extractions of MRI brain scans, namely, the QELBP and the deep learning DL features. The classification of MRI brain scan is improved by exploiting the excellent performance of the QELBP-DL feature extraction of the brain in MRI brain scans. The combining all of the extracted features increase the classification accuracy of long short-term memory network when using it as the brain tumor classifier. The maximum accuracy achieved for classifying a dataset comprising 154 MRI brain scan is 98.80%. The experimental results demonstrate that combining the extracted features improves the performance of MRI brain tumor classification.
There is an increased interest in the study of eye movements during the retrieval of autobiographical memories. Following this trend, the aim of the current study was to evaluate eye movements during the retrieval of remote and recent autobiographical memories. We instructed 71 participants to retrieve memories of personal events from early childhood (6-10 years), late childhood/early adolescence (11-14 years), late adolescence (15-18 years), and the last month. During the retrieval of these memories, participants wore eye-tracking glasses. Analyses showed that early childhood memories triggered fewer fixations and fixations with longer durations than memories from the last month. However, no significant differences were observed for the number of saccades, saccade durations, or total amplitude of the saccades. The fewer and longer lasting fixations during the retrieval of early childhood memories can be attributed either to the visual system reconstructing remote memories from an observer perspective or to difficulties when reconstructing remote memories.
Introduction: There are increasing trend in using information and communication technology to enhance the deliverance of reminiscence work for people with dementia. Thus this study aimed to explore the feasibility of Digital Memory Album (DMA) to support reminiscence work and subsequently evaluate the psychosocial benefits of the DMA system for older adults with dementia living in community. Method: This was an exploratory case study involved five participants with mild to moderate dementia and their primary caregivers. Each participant had an opportunity to review their own life history using Life Review Experience Form (LREF) with the researcher for 8 consecutive weeks in which subsequently facilitated in development of a personalised digital life story. Caregivers provided additional information to support the development of DMA. The whole process was audio recorded. The digital life story was presented in multimedia format and displayed using the DMA. The DMA was given to the participants for review. Qualitative data was collected using semi structure questions with the participants and caregivers immediately after the completion of DMA and 6 weeks after having the DMA as a gift. Results: Analysis of the audio-records and interview data indicated that DMA promoted psychosocial wellbeing which included enjoyment, comfort, stimulate long term memory and enhanced communication and social interactions with family members and friends. Caregivers expressed the DMA helped them in recollecting the past and better understanding of their loved ones. It also reoriented their focus on their loved ones from their disability to the remaining strength. Conclusion: For the first time, this study demonstrates the feasibility of using DMA in improving the psychosocial wellbeing for people with dementia in Malaysia.
Using steady-state electrical properties of non-ohmic dendrite based on cable theory, we derive electrotonic potentials that do not change over time and are localized in space. We hypothesize that clusters of such stationary, local and permanent pulses are the electrical signatures of enduring memories which are imprinted through nonsynaptic plasticity, encoded through epigenetic mechanisms, and decoded through electrotonic processing. We further hypothesize how retrieval of an engram is made possible by integration of these permanently imprinted standing pulses in a neural circuit through neurotransmission in the extracellular space as part of conscious recall that acts as a guiding template in the reconsolidation of long-term memories through novelty characterized by uncertainty that arises when new fragments of memories reinstate an engram by way of nonsynaptic plasticity that permits its destabilization. Collectively, these findings seem to reinforce this hypothesis that electrotonic processing in non-ohmic dendrites yield insights into permanent electrical signatures that could reflect upon enduring memories as fragments of long-term memory engrams.
This study introduces an augmented Long-Short Term Memory (LSTM) neural network architecture, integrating Symbolic Genetic Programming (SGP), with the objective of forecasting cross-sectional price returns across a comprehensive dataset comprising 4500 listed stocks in the Chinese market over the period from 2014 to 2022. Using the S&P Alpha Pool Dataset for China as basic input, this architecture incorporates data augmentation and feature extraction techniques. The result of this study demonstrates significant improvements in Rank Information coefficient (Rank IC) and IC information ratio (ICIR) by 1128% and 5360% respectively when it is applied to fundamental indicators. For technical indicators, the hybrid model achieves a 206% increase in Rank IC and an impressive surge of 2752% in ICIR. Furthermore, the proposed hybrid SGP-LSTM model outperforms major Chinese stock indexes, generating average annualized excess returns of 31.00%, 24.48%, and 16.38% compared to the CSI 300 index, CSI 500 index, and the average portfolio, respectively. These findings highlight the effectiveness of SGP-LSTM model in improving the accuracy of cross-sectional stock return predictions and provide valuable insights for fund managers, traders, and financial analysts.
Preventive maintenance activities require a tool to be offline for long hour in order to perform the prescribed maintenance activities. Although preventive maintenance is crucial to ensure operational reliability and efficiency of the tool, long hour of preventive maintenance activities increases the cycle time of the semiconductor fabrication foundry (Fab). Therefore, this activity is usually performed when the incoming Work-in-Progress to the equipment is forecasted to be low. The current statistical forecasting approach has low accuracy because it lacks the ability to capture the time-dependent behavior of the Work-in-Progress. In this paper, we present a forecasting model that utilizes machine learning method to forecast the incoming Work-In-Progress. Specifically, our proposed model uses LSTM to forecast multistep ahead incoming Work-in-Progress prediction to an equipment group. The proposed model's prediction results were compared with the results of the current statistical forecasting method of the Fab. The experimental results demonstrated that the proposed model performed better than the statistical forecasting method in both hit rate and Pearson's correlation coefficient, r.
Introduction: Just-in-Time Teaching (JiTT) is a novel method of teaching-learning used in various disciplines of science and humanities. It is a technique that scaffolds the students learning process by reducing the cognitive load which is the load related to the executive control of working memory. Just in time teaching underpins the concept that expertise in learning originates from the knowledge already stored in long term memory and with subsequent interaction between the learner and teacher motivates enhanced active learning and optimizes the intellectual performance.
Objective: The objective of the study is to assess the acceptability of JiTT over traditional lecture teaching among the medical students.
Method: In a cross sectional pilot study, three hundred thirty five students participated in a questionnaire based study that grades the various aspects of the traditional lecture series. After the implementation of JiTT method for a period of four months, three hundred two students completed the questionnaire, grading the same variables for JiTT. The analysis was done by using SPSS version 11 applying Paired t test and McNemar's Test.
Result: The results of our study highlighted that JiTT technique was perceived superior to traditional teaching with statistically significant outcomes in the clarity of the topic (p=0.003,) duration of the session (p=0.002), knowledge gained and orientation for exams (p=0.044). The students perceived JiTT method as less monotonous (p=0.005) increasing their alertness during these sessions (p=0.002).
Conclusion: We therefore propose that Just-in-Time Teaching method is a more interactive and acceptable teaching-learning tool shifting the nature of teaching to a more student-centric approach as perceived by the medical students. This is the first pioneer study on JiTT to be performed on undergraduate medical students so far.
Memories associated with substance use disorders, or substance-associated cues increase the likelihood of craving and relapse during abstinence. There is a growing consensus that manipulation of synaptic plasticity may reduce the strength of substance abuse-related memories. On the biological front, there are new insights that suggest memories associated with substance use disorder may follow unique neurobiological pathways that render them more accessible to pharmacological intervention. In parallel to this, research in neurochemistry has identified several potential candidate molecules that could influence the formation and maintenance of long-term memory. Drugs that target these molecules (blebbistatin, isradipine and zeta inhibitory peptide) have shown promise at the preclinical stage. In this review, we shall provide an overview of the evolving understanding on the biochemical mechanisms involved in memory formation and expound on the premise that substance use disorder is a learning disorder.
The fixed oil of black cumin seeds, Nigella sativa L. (NSO), has shown considerable antioxidant and anti-inflammatory activities. Chronic cerebral hypoperfusion has been linked to neurodegenerative disorders including Alzheimer's disease (AD) and its subsequent cognitive impairment in which oxidative stress and neuroinflammation are the principal culprits. Cerebrovascular hypoperfusion was experimentally achieved by bilateral common carotid arteries occlusion (2VO) in rats. Morris water maze (MWM) test was employed to assess the effects of NSO on spatial cognitive function before and after 2VO intervention. Rats were divided into long-term memory (LTM) and short-term memory (STM) groups, each was further subdivided into 3 subgroups: sham control, untreated 2VO and NSO treated 2VO group. All subgroups were tested with MWM at the tenth postoperative week. Working memory test results for both sham control and NSO treated groups showed significantly lower escape latency time and total distance travelled than untreated 2VO group. Similarly, LTM and STM MWM tests for sham control and NSO treated groups revealed significantly better maze test performance as compared to untreated 2VO group. Sham control and NSO treated 2VO groups demonstrated superior probe memory test performance as compared to untreated 2VO group. The fixed oil of Nigella sativa seeds has demonstrated noticeable spatial cognitive preservation in rats challenged with chronic cerebral hypoperfusion which indicates a promising prospective neuroprotective effect.
Memory impairment (MI) is one of the predominant criteria generally used to identify schizophrenia, dementia and amnesia that are associated with neurodegenerative disorders by evaluating patient's cognitive symptoms. To date, there is no available treatment that can completely mitigate MI. Currently, there is a trend in recent investigations towards symptomatic therapy approaches using a variety of natural compounds. Mangiferin is one of them that have been investigated extensively. Mangiferin is a naturally occurring potent glucoxilxanthone and is mainly isolated from the Mangifera indica (Mango) plant. This review is aimed at providing a comprehensive overview on the efficacy of mangiferin on MI, based on in-vivo animal studies. After screening through articles identified from Scopus and PubMed based on the inclusion and exclusion criteria, a total of 11 articles between 2009 and 2019 were included. The minimum and maximum dose of mangiferin were 10 and 200 mg/kg respectively and administered over the period of 12-154 days. The results of 11 articles showed that mangiferin effectively improved spatial recognition, episodic aversive events, short- and long-term memories primarily occurring via its antioxidant and anti-inflammatory effects. The outcomes of the review revealed that mangiferin improves memory and cognitive impairment in different animal models, indicating that it has potential preventive and therapeutic roles in MI.
Alzheimer's disease (AD) is a chronic neurodegenerative brain disease which is characterized by impairment in cognitive functioning. Orthosiphon stamineus (OS) Benth. (Lamiaceae) is a medicinal plant found around Southeast Asia that has been employed as treatments for various diseases. OS extract contains many active compounds that have been shown to possess various pharmacological properties whereby in vitro studies have demonstrated neuroprotective as well as cholinesterase inhibitory effects. This study, therefore aimed at determining whether this Malaysian plant derived flavonoid can reverse scopolamine induced learning and memory dysfunction in the novel object recognition (NOR) test and the elevated plus maze (EPM) test. In the present study, rats were treated once daily with OS 50 mg/kg, 100 mg/kg, 200 mg/kg and donepezil 1 mg/kg via oral dosing and were given intraperitoneal (ip) injection of scopolamine 1 mg/kg daily to induce cognitive deficits. Rats were subjected to behavioral analysis to assess learning and memory functions and hippocampal tissues were extracted for gene expression and immunohistochemistry studies. All the three doses demonstrated improved scopolamine-induced impairment by showing shortened transfer latency as well as the higher inflexion ratio when compared to the negative control group. OS extract also exhibited memory-enhancing activity against chronic scopolamine-induced memory deficits in the long-term memory novel object recognition performance as indicated by an increase in the recognition index. OS extract was observed to have modulated the mRNA expression of CREB1, BDNF, and TRKB genes and pretreatment with OS extract were observed to have increased the immature neurons against hippocampal neurogenesis suppressed by scopolamine, which was confirmed by the DCX-positive stained cells. These research findings suggest that the OS ethanolic extract demonstrated an improving effect on memory and hence could serve as a potential therapeutic target for the treatment of neurodegenerative diseases like AD.