A fuzzy-based prediction approach for blood delivery using machine learning and genetic algorithm

Multiple diseases require a blood transfusion on daily basis. The process of a blood transfusion is successful when the type and amount of blood is available and when the blood is transported at the right time from the blood bank to the operating room. Blood distribution has a large portion of the cost in hospital logistics. The blood bank can serve various hospitals; however, amount of blood is limited due to donor shortage. The transportation must handle several requirements such as timely delivery, vibration avoidance, temperature maintenance, to keep the blood usable. In this paper, we discuss in first section the issues with blood delivery and constraint. The second section present routing and scheduling system based on artificial intelligence to deliver blood from the blood-banks to hospitals based on single blood bank and multiple blood banks with respect of the vehicle capacity used to deliver the blood and creating the shortest path. The third section consist on solution for predicting the blood needs for each hospital based on transfusion history using machine learning and fuzzy logic. The last section we compare the results of well-known solution with our solution in several cases such as shortage and sudden changes.

A Spatial-Contextual Indoor Trajectory Prediction Approach via Hidden Markov Models

Unlike outdoor trajectory prediction that has been studied many years, predicting the movement of a large number of users in indoor space like shopping mall has just been a hot and challenging issue due to the ubiquitous emerging of mobile devices and free Wi-Fi services in shopping centers in recent years. Aimed at solving the indoor trajectory prediction problem, in this paper, a hybrid method based on Hidden Markov approach is proposed. The proposed approach clusters Wi-Fi access points according to their similarities first; then, a frequent subtrajectory based HMM which captures the moving patterns of users has been investigated. In addition, we assume that a customer’s visiting history has certain patterns; thus, we integrate trajectory prediction with shop category prediction into a unified framework which further improves the predicting ability. Comprehensive performance evaluation using a large-scale real dataset collected between September 2012 and October 2013 from over 120,000 anonymized, opt-in consumers in a large shopping center in Sydney was conducted; the experimental results show that the proposed method outperforms the traditional HMM and perform well enough to be usable in practice.

Predicting dispensing errors in community pharmacies: An application of the Systematic Human Error Reduction and Prediction Approach (SHERPA)

Introduction The objective of this study was to use a prospective error analysis method to examine the process of dispensing medication in community pharmacy settings and identify remedial solutions to avoid potential errors, categorising them as strong, intermediate, or weak based on an established patient safety action hierarchy tool. Method Focus group discussions and non-participant observations were undertaken to develop a Hierarchical Task Analysis (HTA), and subsequent focus group discussions applied the Systematic Human Error Reduction and Prediction Approach (SHERPA) focusing on the task of dispensing medication in community pharmacies. Remedial measures identified through the SHERPA analysis were then categorised as strong, intermediate, or weak based on the Veteran Affairs National Centre for Patient Safety action hierarchy. Non-participant observations were conducted at 3 pharmacies, totalling 12 hours, based in England. Additionally, 7 community pharmacists, with experience ranging from 8 to 38 years, participated in a total of 4 focus groups, each lasting between 57 to 85 minutes, with one focus group discussing the HTA and three applying SHERPA. A HTA was produced consisting of 10 sub-tasks, with further levels of sub-tasks within each of them. Results Overall, 88 potential errors were identified, with a total of 35 remedial solutions proposed to avoid these errors in practice. Sixteen (46%) of these remedial measures were categorised as weak, 14 (40%) as intermediate and 5 (14%) as strong according to the Veteran Affairs National Centre for Patient Safety action hierarchy. Sub-tasks with the most potential errors were identified, which included ‘producing medication labels’ and ‘final checking of medicines’. The most common type of error determined from the SHERPA analysis related to omitting a check during the dispensing process which accounted for 19 potential errors. Discussion This work applies both HTA and SHERPA for the first time to the task of dispensing medication in community pharmacies, detailing the complexity of the task and highlighting potential errors and remedial measures specific to this task. Future research should examine the effectiveness of the proposed remedial solutions to improve patient safety.

FAILURES PREDICTION APPROACH IN AGILE SOFTWARE DEVELOPMENT

Software failure prediction is an important activity during agile software development as it can help managers to identify the failure modules. Thus, it can reduce the test time, cost and assign testing resources efficiently. RapidMiner Studio9.4 has been used to perform all the required steps from preparing the primary data to visualizing the results and evaluating the outputs, as well as verifying and improving them in a unified environment. Two datasets are used in this work, the results for the first one indicate that the percentage of failure to predict the time used in the test is for all 181 rows, for all test times recorded, is 3% for Mean time between failures (MTBF). Whereas, SVM achieved a 97% success in predicting compared to previous work whose results indicated that the use of Administrative Delay Time (ADT) achieved a statistically significant overall success rate of 93.5%. At the same time, the second dataset result indicates that the percentage of failure to predict the time used is 1.5% for MTBF, SVM achieved 98.5% prediction.

Deep neural learning based protein function prediction

<abstract> <p>It is vital for the annotation of uncharacterized proteins by protein function prediction. At present, Deep Neural Network based protein function prediction is mainly carried out for dataset of small scale proteins or Gene Ontology, and usually explore the relationships between single protein feature and function tags. The practical methods for large-scale multi-features protein prediction still need to be studied in depth. This paper proposes a DNN based protein function prediction approach IGP-DNN. This method uses Grasshopper Optimization Algorithm (GOA) and Intuitionistic Fuzzy c-Means clustering (IFCM) based protein function modules extracting algorithm to extract the features of protein modules, utilizing Kernel Principal Component Analysis (KPCA) method to reduce the dimensionality of the protein attribute information, and integrating module features and attribute features. Inputting integrated data into DNN through multiple hidden layers to classify proteins and predict protein functions. In the experiments, the F-measure value of IGP-DNN on the DIP dataset reaches 0.4436, which shows better performance.</p> </abstract>

A New Prediction Approach for Preventing Default Customers from Applying Personal Loans Using Machine Learning

In the Egyptian banking industry, loan officers use pure judgment to make personal loan approval decisions. In this paper, we develop a new predictive method for default customers' loans using machine learning. The new predictive method uses the available personal data and historical credit data to evaluate the credit trust-worthiness of customers to obtain loans. We used the ABE dataset for training and testing, as we used 10 features from the application form and i- score report class that could give great help to credit officers for taking the right decision through avoiding customer selection using random techniques. The collected dataset was analysed by using various machine learning classifiers based on important selected features, to obtain high accuracy. We compared the performance of several machine learning classifiers before and after feature selection. We have found that in terms of high accuracy, the most important features are (activity – income – loan) and in terms of better performance the decision tree classifier has surpassed any other machine learning classifier with significant prediction accuracy of almost 94.85%.

A federated learning based semi-supervised credit prediction approach enhanced by multi-layer label mean

Learning based credit prediction has attracted great interest from academia and industry. Different institutions hold a certain amount of credit data with limited users to build model. An institution has the requirement to obtain data from other institutions for improving model performance. However, due to the privacy protection and subject to legal restrictions, they encounter difficulties in data exchange. This affects the performance of credit prediction. In order to solve the above problem, this paper proposes a federated learning based semi-supervised credit prediction approach enhanced by multi-layer label mean, which can aggregate parameters of each institution via joint training while protecting the data privacy of each institution. Moreover, in actual production and life, there are usually more unlabeled credit data than labeled ones, and the distribution of their feature space presents multiple data-dense divisions. To deal with these, local meanNet model is proposed with a multi-layer label mean based semi-supervised deep learning network. In addition, this paper introduces a cost-sensitive loss function in the supervised part of the local mean model. Conducted on two public credit datasets, experimental results show that our proposed federated learning based approach has achieved promising credit prediction performance in terms of Accuracy and F1 measures. At the same time, the framework design mode that splits data aggregation and keys uniformly can improve the security of data privacy and enhance the flexibility of model training.