This study proposes an architecture based on a weighted fuzzy Dempster-Shafer framework (WFDSF), which can adjust weights associated with inconsistent evidence obtained by different classification approaches, to realize a fusion system for integrating multimodal information. The Dempster-Shafer theory (D-S theory) of evidence enables us to integrate heterogeneous information from multiple sources to obtain collaborative inferences for a given problem. To conquer various uncertainties associated with the collected information, our system assigns beliefs and plausibilities to possible hypotheses of each decision maker and uses a combination rule to fuse multimodal information. For information fusion, an important step in D-S aggregation is to find an appropriate basic probability assignment scheme for allocating support to each possible hypothesis/class, which remains an arduous and unsolved problem. Here, we propose a mathematical structure to aggregate weighted evidence extracted from two different types of approaches: fuzzy Naïve Bayes and nearest mean classification rule. Further, an intuitionistic belief assignment is employed to address uncertainties between hypotheses/classes. Finally, 12 benchmark problems from the UCI machine learning repository for classification are employed to validate the proposed WFDSF-based scheme. In addition, an application of WFDSF to a practical brain-computer interface problem involving multimodal data fusion is demonstrated in this study. The experimental results show that the WFDSF is superior to several existing methods.
Study Objectives: Polysomnography is the gold standard for diagnosis of obstructive sleep apnea (OSA) but it is costly and access is often limited. The aim of this study is to develop a clinically useful support vector machine (SVM)-based prediction model to identify patients with high probability of OSA for nonsleep specialist physician in clinical practice.Methods: The SVM model was developed using the features routinely collected at the clinical evaluation from 6,875 Chinese patients referred to sleep clinics for suspected OSA. Three apnea-hypopnea index (AHI) cutoffs, ≥5/h, ≥15/h, and ≥30/h were used to define the severity of OSA. The continuous and categorized features were selected separately and were further selected through stepwise forward feature selection. The modeling was achieved through fivefold cross-validation. The model discriminative ability was evaluated for the whole data set and four subgroups categorized with gender and age (<65 versus ≥65 years old [y/o]).Results: Two features were selected to predict AHI cutoff ≥5/h with six features selected for ≥15/h, and six features selected for ≥30/h, respectively, to reach Area under the Receiver Operating Characteristic (AUROC) 0.82, 0.80, and 0.78, respectively. The sensitivity was 74.14%, 75.18%, and 70.26%, while the specificity was 74.71%, 68.73%, and 70.30%, respectively. Compared to logistic regression, Berlin questionnaire, NoSAS Score, and Supersparse Linear Integer Model (SLIM) scoring system, the SVM model performs better with a more balanced sensitivity and specificity. The discriminative ability was best for male <65 y/o and modest for female ≥65 y/o.Conclusion: Our model provides a simple and accurate modality for early identification of patients with OSA and may potentially help prioritize them for sleep study.
12 December 2017 in IEEE Transactions on Knowledge and Data Engineering
The class imbalance problem in machine learning occurs when certain classes are underrepresented relative to the others, leading to a learning bias toward the majority classes. To cope with the skewed class distribution, many learning methods featuring minority oversampling have been proposed, which are proved to be effective. To reduce information loss during feature space projection, this study proposes a novel oversampling algorithm, named minority oversampling in kernel adaptive subspaces (MOKAS), which exploits the invariant feature extraction capability of a kernel version of the adaptive subspace self-organizing maps. The synthetic instances are generated from well-trained subspaces and then their pre-images are reconstructed in the input space. Additionally, these instances characterize nonlinear structures present in the minority class data distribution and help the learning algorithms to counterbalance the skewed class distribution in a desirable manner. Experimental results on both real and synthetic data show that the proposed MOKAS is capable of modeling complex data distribution and outperforms a set of state-of-the-art oversampling algorithms.
12 October 2017 in IEEE Systems, Man, and Cybernetics Magazine
This article presents the latest BCI-related research done in our group. Our previous work applied computational intelligence technology in BCIs to inspire detailed investigations of practical issues in real-life applications. Novel EEG devices featuring dry electrodes facilitate and speed up electrode positioning before recording and allow subjects to move freely in operational environments. We also demonstrate the feasibility of applying CCA, RBFNs, effective connectivity measurements, and D-S theory to help BCIs extract informative knowledge from brain signals. Two recent trends in research in the computational and artificial intelligence community, big data and deep learning, are expected to impact the direction and development of BCIs.
Background: Knowledge graph embedding is an effective semantic representation method for entities and relations in knowledge graphs. Several translation-based algorithms, including TransE, TransH, TransR, TransD, and TranSparse, have been proposed to learn effective embedding vectors from typical knowledge graphs in which the relations between head and tail entities are deterministic. However, in medical knowledge graphs, the relations between head and tail entities are inherently probabilistic. This difference introduces a challenge in embedding medical knowledge graphs. Objective: We aimed to address the challenge of how to learn the probability values of triplets into representation vectors by making enhancements to existing TransX (where X is E, H, R, D, or Sparse) algorithms, including the following: (1) constructing a mapping function between the score value and the probability, and (2) introducing probability-based loss of triplets into the original margin-based loss function. Methods: We performed the proposed PrTransX algorithm on a medical knowledge graph that we built from large-scale real-world electronic medical records data. We evaluated the embeddings using link prediction task. Results: Compared with the corresponding TransX algorithms, the proposed PrTransX performed better than the TransX model in all evaluation indicators, achieving a higher proportion of corrected entities ranked in the top 10 and normalized discounted cumulative gain of the top 10 predicted tail entities, and lower mean rank. Conclusions: The proposed PrTransX successfully incorporated the uncertainty of the knowledge triplets into the embedding vectors.