Advances in computational intelligence最新文献

Finding the definitions of acronyms in any given text has been an on going problem with multiple proposed solutions. In this paper, we use the bidirectional encoder representations from transformers question answer model provided by Google to find acronym definitions in a given text. Given an acronym and a passage containing the acronym, our model is expected to find the expansion of the acronym in the passage. Through our experiments, we show that this model can correctly predict 94% of acronym expansions assuming a Jaro–Winkler threshold distance of greater than 0.8. One of the main contributions of this paper is a systematic method to create datasets and use them to build a corpus for acronym expansion. Our approach for data generation can be used in many applications where there are no standard datasets.

Type 2 diabetes has recently acquired the status of an epidemic silent killer, though it is non-communicable. There are two main reasons behind this perception of the disease. First, a gradual but exponential growth in the disease prevalence has been witnessed irrespective of age groups, geography or gender. Second, the disease dynamics are very complex in terms of multifactorial risks involved, initial asymptomatic period, different short-term and long-term complications posing serious health threat and related co-morbidities. Majority of its risk factors are lifestyle habits like physical inactivity, lack of exercise, high body mass index (BMI), poor diet, smoking except some inevitable ones like family history of diabetes, ethnic predisposition, ageing etc. Nowadays, machine learning (ML) is increasingly being applied for alleviation of diabetes health burden and many research works have been proposed in the literature to offer clinical decision support in different application areas as well. In this paper, we present a review of such efforts for the prevention and management of type 2 diabetes. Firstly, we present the medical gaps in diabetes knowledge base, guidelines and medical practice identified from relevant articles and highlight those that can be addressed by ML. Further, we review the ML research works in three different application areas namely—(1) risk assessment (statistical risk scores and ML-based risk models), (2) diagnosis (using non-invasive and invasive features), (3) prognosis (from normoglycemia/prior morbidity to incident diabetes and prognosis of incident diabetes to related complications). We discuss and summarize the shortcomings or gaps in the existing ML methodologies for diabetes to be addressed in future. This review provides the breadth of ML predictive modeling applications for diabetes while highlighting the medical and technological gaps as well as various aspects involved in ML-based diabetes clinical decision support.

Machine-learning techniques are being used in the health-care industry to improve care delivery at a lower cost and in less time. Artificial Neural Network (ANN) is well-known machine-learning techniques for its diagnostic applications, but it is also increasingly being utilized to guide health-care management decisions. At the same time, in the healthcare industry, ANN has made significant progress in solving a variety of real-world classification problems that range from linear to non-linear and also from simple to complex. In this research work, an Adaptive Hybrid Classification Model named as Genetic-based Linear Adaptive Skipping Training (GLAST) Algorithm has been proposed for the health-care dataset. It has been designed as two-stage process. In first stage, Genetic Algorithm (GA) is adapted to optimize the Learning rate. After optimizing the Learning rate, the optimal Learning rate has been set to the ANN model is ŋ = 1e−4. In the second stage, The training process is carried out using the Linear Adaptive Skipping Training (LAST) algorithm, which reduces the total training time and thus increases the training speed. As a result, the highlighted characteristics of LAST have been integrated with GA to accomplish rapid classification and enhance computational efficiency. On 8 different health-care datasets extracted from the UCI Repository, the proposed GLAST algorithm outperforms both the BPN and LAST algorithms in terms of accuracy and training time, according to simulation results. The result analyses have proved that the efficiency of this proposed GLAST Algorithm outperforms over the existing techniques such as BPN and LAST in terms of accuracy and training time. On various datasets, experimental results show that GLAST improves accuracy from 4 to 17% over BPN training algorithm and reduces overall training time from 10 to 57% over BPN training algorithm.

The increasing data volumes impose unprecedented challenges to traditional data mining in data preprocessing, learning, and analyzing, it has attracted much attention in designing efficient compressing, indexing and searching methods recently. Inspired by locally sensitive hashing (LSH), divide-and-conquer strategy, and double-voting mechanism, we proposed an iterative instance selection algorithm, which needs to run p rounds iteratively to reduce or eliminate the unwanted bias of the optimal solution by double-voting. In each iteration, the proposed algorithm partitions the big dataset into several subsets and distributes them to different computing nodes. In each node, the instances in local data subset are transformed into Hamming space by l hash function in parallel, and each instance is assigned to one of the l hash tables by the corresponding hash code, the instances with the same hash code are put into the same bucket. And then, a proportion of instances are randomly selected from each hash bucket in each hash table, and a subset is obtained. Thus, totally l subsets are obtained, which are used for voting to select the locally optimal instance subset. The process is repeated p times to obtain p subsets. Finally, the globally optimal instance subset is obtained by voting with the p subsets. The proposed algorithm is implemented with two open source big data platforms, Hadoop and Spark, and experimentally compared with three state-of-the-art methods on testing accuracy, compression ratio, and running time. The experimental results demonstrate that the proposed algorithm provides excellent performance and outperforms three baseline methods.

Aiming at the problems of single input mode and lack of naturalness in the assembly process of existing AR systems, a tracking registration method of mobile AR assembly system is proposed based on multi-quantity and multi-degree of freedom natural fingertip interaction. Firstly, the real-time and stable tracking of hand area in complex environment is realized based on the hand region tracking; secondly, the fingertip detection and recognition based on K-COS and parallel vector is used to improve the precision and stability of fingertip recognition; thirdly, the special movement track of fingertip is recognized based on improved DTW algorithm, which has strong compatibility and feature gradient transformation for complex fingertip trajectory recognition; finally, through the real-time transformation of projection relationship between fingertip and virtual object, the interaction between fingertip and virtual object is made more natural and realistic. The experimental results show that in the complex environment of background, illumination, scale and rotation, the precision of fingertip detection and recognition is about 93%, and the precision of fingertip motion template matching is about 91%. The translation error of the registration method based on visual feature recognition is reduced by about 100pix compared with fingertip tracking registration method, and the efficiency of mobile AR-guided assembly method is improved by about 24.77% compared with the traditional manual assisted assembly method. These data verifies the strong interaction and practicability of the fingertips based on the user's multi-quantity and multi-degree of freedom features in the process of mobile AR guided assembly.

The expansion of Internet-of-Things and Information and Communication Technology allows the application of intelligent concepts to university campus spaces. Several Smart Campus models were implemented recently. However, solutions that foster flexibility in the incorporation of new hardware and software solutions on the existing infrastructure are still a gap, motivating this research. Sapientia smart campus model promotes flexibility by facilitating the incorporation of new solutions on existing infrastructure. The model’s architecture is composed of layers that facilitate technology management and update. A university campus received a model implementation, allowing the execution of experiments to evaluate the incorporation of new hardware and applications. These included a mobile application to support user orientation and internal applications to collect and process campus information, such as temperature. The experiments show how the model incorporates a new hardware component and two new applications on the existing infrastructure. Also, it evidenced the use of the installed devices in more than one application with distinct configurations and purposes.

This systematic review focuses on control strategies and machine learning techniques used in prosthetic knees for restoring mobility of individuals with trans-femoral amputations. Review and classification of control strategies that determine how these prosthetic knees interact with the user and gait strategy inspired algorithms for phase identification, locomotion mode, and motion intention recognition were studied. Relevant studies were identified using electronic databases such as PubMed, EMBASE, SCOPUS, and the Cochrane Controlled Trials Register (Rehabilitation and Related Therapies) up to April 2021. Abstracts were screened and inclusion and exclusion criteria were applied. Out of 278 potentially relevant studies, 65 articles were included. The specific variables on control approach, control modes, gait control, hardware level, machine learning algorithm, and measured signals mechanism were extracted and added to summary table. The results indicate that advanced methods for adapting position or torque depiction and automatic detection of terrains or gait modes are more commonly utilized, but they are largely limited to laboratory environments. It is concluded that a correct combination of control strategies and machine learning techniques will enable the improvement of prosthetic performance and enhance the standard of amputee’s lives.

Many collected data from real world applications often evolve when new attributes or objects are inserted or old ones are removed. The set approximations of ordered information systems (OIS) need to be updated from time to time for further data reduction, analysis, or decision-making. Incremental approaches are feasible and efficient techniques for updating approaches when any variation occurs. In this paper, considering OIS for multi-criteria classification problems, we discuss the principles of incrementally updating approximations in dominance relation based method in four different types of dynamic environments which combine the changes of both attribute set and object set. In each dynamic environment, the corresponding updating principles and algorithm are given with detail proofs. The experimental results and analysis on UCI data sets show that the proposed incremental approach outperforms the non-incremental method and the integration of current incremental algorithms in the implementation efficiency.

The implementation of machine learning techniques allows to prevent in advance the degeneration of any component present in a wind turbine, as well as the detection and diagnosis of sudden failures. This methodology allows automatic and autonomous learning to predict, detect and diagnose electrical and mechanical failures in wind turbines. Four different failure states have been simulated due to bearing vibrations in wind turbines, comparing traditional techniques, such as frequency analysis, as well as the implementation of AI, using the KNN and SVM methodology. This contribution evaluates different methodologies for monitoring, supervision and fault diagnosis based on the implementation of machine learning algorithms adapted to the different components and faults of the wind turbine. Implementing these techniques, allows to anticipate a breakdown, reduce downtime and costs, especially if they are offshore.

Modern search engine result pages (SERPs) become increasingly complex with heterogeneous information aggregated from various sources. In many cases, these SERPs also display results in the right rail besides the traditional left-rail result lists, which change the linear result list to a non-linear panel and might influence user search behavior patterns. While user behavior on the traditional ranked result list has been well studied in existing works, it still lacks a thorough investigation of the effects caused by the right-rail results, especially on complex SERPs. To shed light on this research question, we conducted a user study, which collected participants’ eye movements, detailed interaction behavioral logs, and feedback information. Based on the collected data, we analyze the influence of right-rail results on users’ examination patterns, search behavior, perceived workload, and satisfaction. We further construct a user model to predict users’ examination behavior on non-linear SERPs. Our work contributes to understanding the effects of the right-rail results on users’ interaction patterns, benefiting other related research, such as the evaluation and UI optimization of search systems.