Search

In this work, we have presented a way to increase the contrast of an image. Our target is to find a transformation that will be image specific. We have used a fuzzy system as our transformation function. To tune the system according to an image, we have used Genetic Algorithm and Hill Climbing in multiple ways to evolve the fuzzy system and conducted several experiments. Different variants of the method are tested on several images and two variants that are superior to others in terms of fitness are selected. We have also conducted a survey to assess the visual improvement of the enhancements made by the two variants. The survey indicates that one of the methods can enhance the contrast of the images visually., Comment: 14 pages, 7 figures, Image Processing, Computer Vision, Evolutionary Computation

In this work, we introduce BanglaBERT, a BERT-based Natural Language Understanding (NLU) model pretrained in Bangla, a widely spoken yet low-resource language in the NLP literature. To pretrain BanglaBERT, we collect 27.5 GB of Bangla pretraining data (dubbed `Bangla2B+') by crawling 110 popular Bangla sites. We introduce two downstream task datasets on natural language inference and question answering and benchmark on four diverse NLU tasks covering text classification, sequence labeling, and span prediction. In the process, we bring them under the first-ever Bangla Language Understanding Benchmark (BLUB). BanglaBERT achieves state-of-the-art results outperforming multilingual and monolingual models. We are making the models, datasets, and a leaderboard publicly available at https://github.com/csebuetnlp/banglabert to advance Bangla NLP., Findings of North American Chapter of the Association for Computational Linguistics, NAACL 2022 (camera-ready)

Lung cancer is one of the deadliest cancers, and in part its effective diagnosis and treatment depend on the accurate delineation of the tumor. Human-centered segmentation, which is currently the most common approach, is subject to inter-observer variability, and is also time-consuming, considering the fact that only experts are capable of providing annotations. Automatic and semi-automatic tumor segmentation methods have recently shown promising results. However, as different researchers have validated their algorithms using various datasets and performance metrics, reliably evaluating these methods is still an open challenge. The goal of the Lung-Originated Tumor Segmentation from Computed Tomography Scan (LOTUS) Benchmark created through 2018 IEEE Video and Image Processing (VIP) Cup competition, is to provide a unique dataset and pre-defined metrics, so that different researchers can develop and evaluate their methods in a unified fashion. The 2018 VIP Cup started with a global engagement from 42 countries to access the competition data. At the registration stage, there were 129 members clustered into 28 teams from 10 countries, out of which 9 teams made it to the final stage and 6 teams successfully completed all the required tasks. In a nutshell, all the algorithms proposed during the competition, are based on deep learning models combined with a false positive reduction technique. Methods developed by the three finalists show promising results in tumor segmentation, however, more effort should be put into reducing the false positive rate. This competition manuscript presents an overview of the VIP-Cup challenge, along with the proposed algorithms and results.

Data transformation (DT) is a process that transfers the original data into a form which supports a particular classification algorithm and helps to analyze the data for a special purpose. To improve the prediction performance we investigated various data transform methods. This study is conducted in a customer churn prediction (CCP) context in the telecommunication industry (TCI), where customer attrition is a common phenomenon. We have proposed a novel approach of combining data transformation methods with the machine learning models for the CCP problem. We conducted our experiments on publicly available TCI datasets and assessed the performance in terms of the widely used evaluation measures (e.g. AUC, precision, recall, and F-measure). In this study, we presented comprehensive comparisons to affirm the effect of the transformation methods. The comparison results and statistical test proved that most of the proposed data transformation based optimized models improve the performance of CCP significantly. Overall, an efficient and optimized CCP model for the telecommunication industry has been presented through this manuscript., Comment: 24 pages

In this note, we consider the problem of counting and verifying abelian border arrays of binary words. We show that the number of valid abelian border arrays of length \(n\) is \(2^{n-1}\). We also show that verifying whether a given array is the abelian border array of some binary word reduces to computing the abelian border array of a specific binary word. Thus, assuming the word-RAM model, we present an \(O\left(\frac{n^2}{\log^2n}\right)\) time algorithm for the abelian border array verification problem.

We are presenting COVID-19Base, a knowledgebase highlighting the biomedical entities related to COVID-19 disease based on literature mining. To develop COVID-19Base, we mine the information from publicly available scientific literature and related public resources. We considered seven topic-specific dictionaries, including human genes, human miRNAs, human lncRNAs, diseases, Protein Databank, drugs, and drug side effects, are integrated to mine all scientific evidence related to COVID-19. We have employed an automated literature mining and labeling system through a novel approach to measure the effectiveness of drugs against diseases based on natural language processing, sentiment analysis, and deep learning. To the best of our knowledge, this is the first knowledgebase dedicated to COVID-19, which integrates such large variety of related biomedical entities through literature mining. Proper investigation of the mined biomedical entities along with the identified interactions among those, reported in COVID-19Base, would help the research community to discover possible ways for the therapeutic treatment of COVID-19., Comment: 10 pages, 3 figures

Malaria is a female anopheles mosquito-bite inflicted life-threatening disease which is considered endemic in many parts of the world. This article focuses on improving malaria detection from patches segmented from microscopic images of red blood cell smears by introducing a deep convolutional neural network. Compared to the traditional methods that use tedious hand engineering feature extraction, the proposed method uses deep learning in an end-to-end arrangement that performs both feature extraction and classification directly from the raw segmented patches of the red blood smears. The dataset used in this study was taken from National Institute of Health named NIH Malaria Dataset. The evaluation metric accuracy and loss along with 5-fold cross validation was used to compare and select the best performing architecture. To maximize the performance, existing standard pre-processing techniques from the literature has also been experimented. In addition, several other complex architectures have been implemented and tested to pick the best performing model. A holdout test has also been conducted to verify how well the proposed model generalizes on unseen data. Our best model achieves an accuracy of almost 97.77%., Comment: Application of deep learning in biological science for the early detection of disease

Class imbalance problem is commonly faced while developing machine learning models for real-life issues. Due to this problem, the fitted model tends to be biased towards the majority class data, which leads to lower precision, recall, AUC, F1, G-mean score. Several researches have been done to tackle this problem, most of which employed resampling, i.e. oversampling and undersampling techniques to bring the required balance in the data. In this paper, we propose neural network based algorithms for undersampling. Then we resampled several class imbalanced data using our algorithms and also some other popular resampling techniques. Afterwards we classified these undersampled data using some common classifier. We found out that our resampling approaches outperform most other resampling techniques in terms of both AUC, F1 and G-mean score., Comment: 8 pages in IEEE format

Though the interfacial tractor beam experiment supports the increase of photon momentum (i.e., Minkowski momentum), it is still a matter of investigation whether, inside matter, the photon momentum always increases. Considering the inhomogeneous or heterogeneous background, we have demonstrated that if the background and the half-immersed object are both non-absorbing, the transferred photon momentum to the object can be considered as the one of Minkowski exactly at the interface. In contrast, the presence of loss inside matter, either in the half-immersed object (i.e., plasmonic or lossy dielectric) or in the background, changes the whole situation. For such cases, our several demonstrations and proposed thought experiments have strongly supported the decrease of photon momentum instead of the usual perception of its increase. Although almost all the major radiation pressure experiments have so far supported the linear increase of photon momentum, our proposed simple experimental set-ups may introduce a novel way to observe and verify the exactly opposite proposal: the Abraham momentum of photon. Finally, as an interesting sidewalk, based on several parameters, a machine learning based system has been developed to predict the transferred momentum of photon within a very short time avoiding time-consuming full simulation.

We first describe three algorithms for computing the Lyndon array that have been suggested in the literature, but for which no structured exposition has been given. Two of these algorithms execute in quadratic time in the worst case, the third achieves linear time, but at the expense of prior computation of both the suffix array and the inverse suffix array of x. We then go on to describe two variants of a new algorithm that avoids prior computation of global data structures and executes in worst-case n log n time. Experimental evidence suggests that all but one of these five algorithms require only linear execution time in practice, with the two new algorithms faster by a small factor. We conjecture that there exists a fast and worst-case linear-time algorithm to compute the Lyndon array that is also elementary (making no use of global data structures such as the suffix array).

$V$-order is a global order on strings related to Unique Maximal Factorization Families (UMFFs), which are themselves generalizations of Lyndon words. $V$-order has recently been proposed as an alternative to lexicographical order in the computation of suffix arrays and in the suffix-sorting induced by the Burrows-Wheeler transform. Efficient $V$-ordering of strings thus becomes a matter of considerable interest. In this paper we present new and surprising results on $V$-order in strings, then go on to explore the algorithmic consequences.

In this paper, we consider the problem of finding a minimum common partition of two strings. The problem has its application in genome comparison. As it is an NP-hard, discrete combinatorial optimization problem, we employ a metaheuristic technique, namely, MAX-MIN ant system to solve this problem. To achieve better efficiency we first map the problem instance into a special kind of graph. Subsequently, we employ a MAX-MIN ant system to achieve high quality solutions for the problem. Experimental results show the superiority of our algorithm in comparison with the state of art algorithm in the literature. The improvement achieved is also justified by standard statistical test.

In this paper, we revisit the much studied problem of Pattern Matching with Swaps (Swap Matching problem, for short). We first present a graph-theoretic model, which opens a new and so far unexplored avenue to solve the problem. Then, using the model, we devise two efficient algorithms to solve the swap matching problem. The resulting algorithms are adaptations of the classic shift-and algorithm. For patterns having length similar to the word-size of the target machine, both the algorithms run in linear time considering a fixed alphabet., Comment: 23 pages, 3 Figures and 17 Tables

The {\em longest common subsequence (LCS)} problem is a classic and well-studied problem in computer science. LCS is a central problem in stringology and finds broad applications in text compression, error-detecting codes and biological sequence comparison. However, in numerous contexts, words represent cyclic sequences of symbols and LCS must be generalized to consider all circular shifts of the strings. This occurs especially in computational biology when genetic material is sequenced form circular DNA or RNA molecules. This initiates the problem of {\em longest common cyclic subsequence (LCCS)} which finds the longest subsequence between all circular shifts of two strings. In this paper, we give an $O(n^2)$ algorithm for solving LCCS problem where $n$ is the number of symbols in the strings., Comment: This paper has been withdrawn by the author due to a crucial error in the proofs