Chronic subdural hematoma pressure measurement and its correlation with clinical characteristics and postoperative neurological outcome

Background: The pressure of the chronic SDH (subdural haemotoma), the age of the patient, preoperative GCS score and midline shift were considered prognostic dependent factors. The study aimed at the significance of the pressure of chronic SDH in the outcome of patients.Methods: A correlation between subdural hematoma pressure and preoperative and postoperative clinical variables such as hematoma volume, midline shift, age, GCS score and postoperative modified ranking scale score as well as complications were assessed and analyzed.Results: According to the pressure of chronic SDH, 56 patients were grouped into 4 groups. In the pressure group <15 cm/h20 group the mean age was 85 and postoperative ranking score was 3 and the recurrence was 21 % while in high pressure group (>25 cm/h20) the mortality was 14% and no recurrence.Conclusions: The pressure of the chronic SDH has significant prognostic value in chronic SDH surgeries.

Multi-Keyword Classification: A Case Study in Finnish Social Sciences Data Archive

In this paper, we consider the task of assigning relevant labels to studies in the social science domain. Manual labelling is an expensive process and prone to human error. Various multi-label text classification machine learning approaches have been proposed to resolve this problem. We introduce a dataset obtained from the Finnish Social Science Archive and comprised of 2968 research studies’ metadata. The metadata of each study includes attributes, such as the “abstract” and the “set of labels”. We used the Bag of Words (BoW), TF-IDF term weighting and pretrained word embeddings obtained from FastText and BERT models to generate the text representations for each study’s abstract field. Our selection of multi-label classification methods includes a Naive approach, Multi-label k Nearest Neighbours (ML-kNN), Multi-Label Random Forest (ML-RF), X-BERT and Parabel. The methods were combined with the text representation techniques and their performance was evaluated on our dataset. We measured the classification accuracy of the combinations using Precision, Recall and F1 metrics. In addition, we used the Normalized Discounted Cumulative Gain to measure the label ranking performance of the selected methods combined with the text representation techniques. The results showed that the ML-RF model achieved a higher classification accuracy with the TF-IDF features and, based on the ranking score, the Parabel model outperformed the other methods.

Strategic Supplier Selection in Payment Industry: A Multi-Criteria Solution for Insufficient and Interrelated Data Sources

Our goal is to address the complicated problem of strategic supplier selection with interrelated and insufficient data. To achieve this goal, we proposed our Strategic Supplier Selection Methodology (SSSM). First, SSSM formulates the enterprise strategies and evaluation criteria. Then, we developed a novel method called Grey Principal Component Analysis-Data Envelopment Analysis (GPCA-DEA) to evaluate suppliers in SSSM. GPCA-DEA overcomes the major disadvantages and limitations of former methodologies (e.g., DEA) while dealing with insufficient and interrelated data. Finally, SSSM applies Multiple Attribute Decision-Making (MADM) methods to select suppliers based on the ranking score. The application of SSSM is illustrated in the payment industry to select Payment Initiation Service Providers (PISP). For the first time, we considered the payment industry-specific criteria in compliance with the latest regulation (PSD2). The Spearman rank correlation statistical test showed that our method (GPCA-DEA used in SSSM) yields more reliable results than a former version of DEA.

A new method for predicting future links in temporal networks based on node influence

Link prediction in temporal networks has always been a hot topic in both statistical physics and network science. Most existing works fail to consider the inner relationship between nodes, leading to poor prediction accuracy. Even though a wide range of realistic networks are temporal ones, few existing works investigated the properties of realistic and temporal networks. In this paper, we address the problem of abstracting individual attributes and propose a adaptive link prediction method for temporal networks based on [Formula: see text]-index to predict future links. The matching degree of nodes is first defined considering both the native influence and the secondary influence of local structure. Then a similarity index is designed using a decaying parameter to punish the snapshots with their occurring time. Experimental results on five realistic temporal networks observing consistent gains of 2–9% AUC in comparison to the best baseline in four networks show that our proposed method outperforms several benchmarks under two standard evaluation metrics: AUC and Ranking score. We also investigate the influence of the free parameter and the definition of matching degree on the prediction performance.

Collaborative filtering based service ranking with invocation histories

Collaborative filtering based recommender systems have been very successful on dealing with the information overload problem and providing personalized recommendations to users. When more and more web services are published online, this technique can also help recommend and select services which satisfy users’ particular Quality of Service (QoS) requirements and preferences. In this thesis, we propose a novel collaborative filtering based service ranking mechanism, in which the invocation and query histories are used to infer the users’ preferences, and user similarity is calculated based on invocations and queries. To overcome some of the inherent problems with the collaborative filtering systems such as the cold start and data sparsity problem, the final ranking score is a combination of the QoS-based matching score and the collaborative filtering based score. The experiment using the simulated data proves the effectiveness of the proposed algorithm.

Collaborative filtering based service ranking with invocation histories

Collaborative filtering based recommender systems have been very successful on dealing with the information overload problem and providing personalized recommendations to users. When more and more web services are published online, this technique can also help recommend and select services which satisfy users’ particular Quality of Service (QoS) requirements and preferences. In this thesis, we propose a novel collaborative filtering based service ranking mechanism, in which the invocation and query histories are used to infer the users’ preferences, and user similarity is calculated based on invocations and queries. To overcome some of the inherent problems with the collaborative filtering systems such as the cold start and data sparsity problem, the final ranking score is a combination of the QoS-based matching score and the collaborative filtering based score. The experiment using the simulated data proves the effectiveness of the proposed algorithm.

Neighborhood Attentional Memory Networks for Recommendation Systems

Deep learning systems have been phenomenally successful in the fields of computer vision, speech recognition, and natural language processing. Recently, researchers have adopted deep learning techniques to tackle collaborative filtering with implicit feedback. However, the existing methods generally profile both users and items directly, while neglecting the similarities between users’ and items’ neighborhoods. To this end, we propose the neighborhood attentional memory networks (NAMN), a deep learning recommendation model applying two dedicated memory networks to capture users’ neighborhood relations and items’ neighborhood relations respectively. Specifically, we first design the user neighborhood component and the item neighborhood component based on memory networks and attention mechanisms. Then, by the associative addressing scheme with the user and item memories in the neighborhood components, we capture the complex user-item neighborhood relations. Stacking multiple memory modules together yields deeper architectures exploring higher-order complex user-item neighborhood relations. Finally, the output module jointly exploits the user and item neighborhood information with the user and item memories to obtain the ranking score. Extensive experiments on three real-world datasets demonstrate significant improvements of the proposed NAMN method over the state-of-the-art methods.

The fuzzy integrated TOPSIS economy ranking (FITER) model for manufacturing technology development in China

With knowledge perspective of industrial technology, in this paper we propose fast ranking score decision making model based on Fuzzy integrated TOPSIS approach to determine economic growth rate of manufacturing industry in China. This research focuses on driving effects of China’s productive service industry on manufacturing technology innovation. The research results show that the manufacturer service industry takes a high level of information diffusion for the manufacturing industry. It transmits a large amount of diverse information through the unconstrained relationship with the manufacturing industry, thereby forming the economic network with proposed Fuzzy integrated TOPSIS economy ranking (FITER) model and improve the development level of the manufacturing industry. We evaluate the performance of proposed FITER model by comparing ranking score of different manufacturing industry with different existing decision making mode and demonstrate that proposed model represent best ranking score in comparison to existing approach. Result from data analysis motivates driving effect of production services on the technological innovation of manufacturing and sub-sectors. It is found that the innovation and technological advancement in the production industry of services takes drive the overall expansion level of the business industry, as well as make the manufacturing industry the strongest.

Prediction of Human-Plasmodium vivax Protein Associations From Heterogeneous Network Structures Based on Machine-Learning Approach

Malaria caused by Plasmodium vivax can lead to severe morbidity and death. In addition, resistance has been reported to existing drugs in treating this malaria. Therefore, the identification of new human proteins associated with malaria is urgently needed for the development of additional drugs. In this study, we established an analysis framework to predict human- P. vivax protein associations using network topological profiles from a heterogeneous network structure of human and P. vivax, machine-learning techniques and statistical analysis. Novel associations were predicted and ranked to determine the importance of human proteins associated with malaria. With the best-ranking score, 411 human proteins were identified as promising proteins. Their regulations and functions were statistically analyzed, which led to the identification of proteins involved in the regulation of membrane and vesicle formation, and proteasome complexes as potential targets for the treatment of P. vivax malaria. In conclusion, by integrating related data, our analysis was efficient in identifying potential targets providing an insight into human-parasite protein associations. Furthermore, generalizing this model could allow researchers to gain further insights into other diseases and enhance the field of biomedical science.

Discovery of Potent Sialic Acid Inhibitors Against the Hemagglutinin of Influenza A Virus

Influenza A virus (IAV) is hard to eradicate due to its genetic drift and reassortment ability. Neuraminidase (NA), frequently the target protein, cleaves the sialic acid (SA) and discharges virions to complete the infectious cycle. However, the increasing use of NA inhibitors aroused drug resistance in recent years. Hemagglutinin (HA), on the opposite, initiates the infectious cycle and sticks virions to cells by connecting to the host SA so that HA might be a tempting target. In this study, HA was chosen for SA-binding site model preparation to screen more than 200,000 compounds by molecular docking method in silico. According to the post-screening analysis by iGemdock and SiMMap, nine of the top twenty compounds based on the ranking score were purchased and evaluated. NSC85561 was initially identified from the compounds through the bioassays. Next, the twelve derivatives of NSC85561 were selected to consolidate the primary results. NSC85561 and two of its derivatives were finally identified as potent HA inhibitors by cell proliferation, plaque reduction, and hemagglutination inhibition assays in vitro. These results suggested that virtual screening may be a powerful tool to concise the compounds from the massive database and reduce the complicated bioassays.