A neuro-fuzzy system is a learning machine that finds the parameters of a fuzzy system using approximate techniques of neural networks. Both neural network and fuzzy system have common features. These can solve problems that there is no mathematical model for them. Adaptive Neuro-Fuzzy Inference System (ANFIS) is an adaptive network that uses supervised learning on learning algorithm. Selecting optimization method in training, to achieve effective results with ANFIS is very important. Heuristics and metaheuristics algorithms attempt to find the best solution out of all possible solutions of an optimization problem. ANFIS training can be based on non-derivative algorithms. Heuristics and metaheuristics are non-derivative algorithms that can lead to better performance in ANFIS training. Most heuristic and metaheuristic algorithms are taken from the behavior of biological systems or physical systems in nature. The newly released Emperor Penguins Colony (EPC) algorithm is a population-based and nature-inspired metaheuristic algorithm. This algorithm has many potential for solving various problems. In this paper an optimized ANFIS based on newly EPC algorithm is proposed. The optimized ANFIS is compared with other non-derivative algorithms on benchmark datasets. Eventually, the proposed algorithm is used to solve the classical inverted pendulum problem. The results show that the proposed ANFIS based on EPC algorithm has less error and better performance than other state-of-the-art algorithms in both training and testing phase.

Answer selection (AS) is a critical subtask of the open-domain question answering (QA) problem. The present paper proposes a method called RLAS-BIABC for AS, which is established on attention mechanism-based long short-term memory (LSTM) and the bidirectional encoder representations from transformers (BERT) word embedding, enriched by an improved artificial bee colony (ABC) algorithm for pretraining and a reinforcement learning-based algorithm for training backpropagation (BP) algorithm. BERT can be comprised in downstream work and fine-tuned as a united task-specific architecture, and the pretrained BERT model can grab different linguistic effects. Existing algorithms typically train the AS model with positive-negative pairs for a two-class classifier. A positive pair contains a question and a genuine answer, while a negative one includes a question and a fake answer. The output should be one for positive and zero for negative pairs. Typically, negative pairs are more than positive, leading to an imbalanced classification that drastically reduces system performance. To deal with it, we define classification as a sequential decision-making process in which the agent takes a sample at each step and classifies it. For each classification operation, the agent receives a reward, in which the prize of the majority class is less than the reward of the minority class. Ultimately, the agent finds the optimal value for the policy weights. We initialize the policy weights with the improved ABC algorithm. The initial value technique can prevent problems such as getting stuck in the local optimum. Although ABC serves well in most tasks, there is still a weakness in the ABC algorithm that disregards the fitness of related pairs of individuals in discovering a neighboring food source position. Therefore, this paper also proposes a mutual learning technique that modifies the produced candidate food source with the higher fitness between two individuals selected by a mutual learning factor. We tested our model on three datasets, LegalQA, TrecQA, and WikiQA, and the results show that RLAS-BIABC can be recognized as a state-of-the-art method.

Aim: To determine and compare the patterns of spinal injury in car occupants. Materials and Methods: Retrospective cross-sectional study enrolling all patients with spinal fracture after car accidents, who were admitted to hospital more than 24 h during 2004–2009. Results: The lumbosacral spine was the most commonly involved region (64.8). Six patients had spinal cord injury (6.6%). The majority of the victims were drivers of the vehicle (86.8%) and remaining were passengers. There was a significant difference in lumbar anatomic region (P = 0.05) and place of accident (P = 0.05) in car occupants' position (P = 0.05). Car rollover was the most common mechanism of spinal fractures. There was a significant difference in lumbar anatomic region (P = 0.05), and two or more associated organ injuries (P ≤ 0.05) in car accident mechanism (P = 0.05). Conclusion: The chance of sustaining serious spine and associated multiple injuries in car accidents is quite high in our today's society. This may be due to the low level of standards for car manufacturing, absence or inadequacy of appropriate safety measures in cars, and poorly designed roads and problems in quality of driving to mention some reasons. Therefore, these victims are prone to significant morbidity and even mortality and need more specific prehospital supportive interventions.