Siamese Long Short-Term Memory Research Articles

Efficient cross-lingual plagiarism detection using bidirectional and auto-regressive transformers

<span lang="EN-US">The pervasive availability of vast online information has fundamentally altered our approach to acquiring knowledge. Nevertheless, this wealth of data has also presented significant challenges to academic integrity, notably in the realm of cross-lingual plagiarism. This type of plagiarism involves the unauthorized copying, translation, ideas, or works from one language into others without proper citation. This research introduces a methodology for identifying multilingual plagiarism, utilizing a pre-trained multilingual bidirectional and auto-regressive transformers (mBART) model for document feature extraction. Additionally, a siamese long short-term memory (SLSTM) model is employed for classifying pairs of documents as either "plagiarized" or "non-plagiarized". Our approach exhibits notable performance across various languages, including English (En), Spanish (Es), German (De), and French (Fr). Notably, experiments focusing on the En-Fr language pair yielded exceptional results, with an accuracy of 98.83%, precision of 98.42%, recall of 99.32%, and F-score of 98.87%. For En-Es, the model achieved an accuracy of 97.94%, precision of 98.57%, recall of 97.47%, and an F-score of 98.01%. In the case of En-De, the model demonstrated an accuracy of 95.59%, precision of 95.21%, recall of 96.85%, and F-score of 96.02%. These outcomes underscore the effectiveness of combining the MBART transformer and SLSTM models for cross-lingual plagiarism detection.</span>

A Siamese deep learning framework for efficient hardware Trojan detection using power side-channel data

Hardware Trojans (HTs) are hidden threats embedded in the circuitry of integrated circuits (ICs), enabling unauthorized access, data theft, operational disruptions, or even physical harm. Detecting Hardware Trojans (HTD) is paramount for ensuring IC security. This paper introduces a novel Siamese neural network (SNN) framework for non-destructive HTD. The proposed framework can detect HTs by processing power side-channel signals without the need for a golden model of the IC. To obtain the best results, different neural network models such as Convolutional Neural Network (CNN), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM) are integrated individually with SNN. These models are trained on the extracted features from the Trojan Power & EM Side-Channel dataset. The results show that the Siamese LSTM model achieved the highest accuracy of 86.78%, followed by the Siamese GRU model with 83.59% accuracy and the Siamese CNN model with 73.54% accuracy. The comparison shows that of the proposed Siamese LSTM is a promising new approach for HTD and outperform the state-of-the-art methods.

Exploring Attentive Siamese LSTM for Low-Resource Text Plagiarism Detection

Abstract Low-resource text plagiarism detection faces a significant challenge due to the limited availability of labeled data for training. This task requires the development of sophisticated algorithms capable of identifying similarities and differences in texts, particularly in the realm of semantic rewriting and translation-based plagiarism detection. In this paper, we present an enhanced attentive Siamese Long Short-Term Memory (LSTM) network designed for Tibetan-Chinese plagiarism detection. Our approach begins with the introduction of translation-based data augmentation, aimed at expanding the bilingual training dataset. Subsequently, we propose a pre-detection method leveraging abstract document vectors to enhance detection efficiency. Finally, we introduce an improved attentive Siamese LSTM network tailored for Tibetan-Chinese plagiarism detection. We conduct comprehensive experiments to showcase the effectiveness of our proposed plagiarism detection framework.

ScLSTM: single-cell type detection by siamese recurrent network and hierarchical clustering

MotivationCategorizing cells into distinct types can shed light on biological tissue functions and interactions, and uncover specific mechanisms under pathological conditions. Since gene expression throughout a population of cells is averaged out by conventional sequencing techniques, it is challenging to distinguish between different cell types. The accumulation of single-cell RNA sequencing (scRNA-seq) data provides the foundation for a more precise classification of cell types. It is crucial building a high-accuracy clustering approach to categorize cell types since the imbalance of cell types and differences in the distribution of scRNA-seq data affect single-cell clustering and visualization outcomes.ResultTo achieve single-cell type detection, we propose a meta-learning-based single-cell clustering model called ScLSTM. Specifically, ScLSTM transforms the single-cell type detection problem into a hierarchical classification problem based on feature extraction by the siamese long-short term memory (LSTM) network. The similarity matrix derived from the improved sigmoid kernel is mapped to the siamese LSTM feature space to analyze the differences between cells. ScLSTM demonstrated superior classification performance on 8 scRNA-seq data sets of different platforms, species, and tissues. Further quantitative analysis and visualization of the human breast cancer data set validated the superiority and capability of ScLSTM in recognizing cell types.

Word Embedding for High Performance Cross-Language Plagiarism Detection Techniques

Academic plagiarism has become a serious concern as it leads to the retardation of scientific progress and violation of intellectual property. In this context, we make a study aiming at the detection of cross-linguistic plagiarism based on Natural language Preprocessing (NLP), Embedding Techniques, and Deep Learning. Many systems have been developed to tackle this problem, and many rely on machine learning and deep learning methods. In this paper, we propose Cross-language Plagiarism Detection (CL-PD) method based on Doc2Vec embedding techniques and a Siamese Long Short-Term Memory (SLSTM) model. Embedding techniques help capture the text's contextual meaning and improve the CL-PD system's performance. To show the effectiveness of our method, we conducted a comparative study with other techniques such as GloVe, FastText, BERT, and Sen2Vec on a dataset combining PAN11, JRC-Acquis, Europarl, and Wikipedia. The experiments for the Spanish-English language pair show that Doc2Vec+SLSTM achieve the best results compared to other relevant models, with an accuracy of 99.81%, a precision of 99.75%, a recall of 99.88%, an f-score of 99.70%, and a very small loss in the test phase.

Bearing remaining useful life prediction using self-adaptive graph convolutional networks with self-attention mechanism

Bearings are commonly used to reduce friction between moving parts. Bearings may fail due to lubrication failure, contamination, corrosion, and fatigue. To prevent bearing failures, it is important to predict the remaining useful life (RUL) of bearings. While many data-driven methods have been introduced, very few studies have considered the correlation of features at different time points, such a correlation could be used to identify and aggregate features at different time points for improving the robustness of predictive models. Moreover, many existing data-driven methods leverage neural networks with recurrent characteristics such as recurrent neural network (RNN) and long short term memory (LSTM). These methods are ineffective in processing long sequences and require longer training time due to the recurrent characteristics. To address these issues, a Siamese LSTM network is firstly introduced to classify degradation stages before predicting the RUL of bearings. Then we introduce a self-adaptive graph convolutional network (SAGCN) along with a self-attention mechanism in order to consider the correlation of features at different time points without using recurrent characteristics. Experimental results have demonstrated that the proposed method can accurately predict the RUL with a minimum average root mean squared error of 0.119, and outperforms existing data-driven methods, such as graph convolutional network, convolutional LSTM, convolutional neural network, and generative adversarial network.

Estimating Demand Flexibility Using Siamese LSTM Neural Networks

There is an opportunity in modern power systems to explore the demand flexibility by incentivizing consumers with dynamic prices. In this paper, we quantify demand flexibility using an efficient tool called time-varying elasticity, whose value may change depending on the prices and decision dynamics. This tool is particularly useful for evaluating the demand response potential and system reliability. Recent empirical evidences have highlighted some abnormal features when studying demand flexibility, such as delayed responses and vanishing elasticities after price spikes. Existing methods fail to capture these complicated features because they heavily rely on some predefined (often over-simplified) regression expressions. Instead, this paper proposes a model-free methodology to automatically and accurately derive the optimal estimation pattern. We further develop a two-stage estimation process with Siamese long short-term memory (LSTM) networks. Here, a LSTM network encodes the price response, while the other network estimates the time-varying elasticities. In the case study, the proposed framework and models are validated to achieve higher overall estimation accuracy and better description for various abnormal features when compared with the state-of-the-art methods.

Comparative Study of Markerless Vision-Based Gait Analyses for Person Re-Identification

The model-based gait analysis of kinematic characteristics of the human body has been used to identify individuals. To extract gait features, spatiotemporal changes of anatomical landmarks of the human body in 3D were preferable. Without special lab settings, 2D images were easily acquired by monocular video cameras in real-world settings. The 2D and 3D locations of key joint positions were estimated by the 2D and 3D pose estimators. Then, the 3D joint positions can be estimated from the 2D image sequences in human gait. Yet, it has been challenging to have the exact gait features of a person due to viewpoint variance and occlusion of body parts in the 2D images. In the study, we conducted a comparative study of two different approaches: feature-based and spatiotemporal-based viewpoint invariant person re-identification using gait patterns. The first method is to use gait features extracted from time-series 3D joint positions to identify an individual. The second method uses a neural network, a Siamese Long Short Term Memory (LSTM) network with the 3D spatiotemporal changes of key joint positions in a gait cycle to classify an individual without extracting gait features. To validate and compare these two methods, we conducted experiments with two open datasets of the MARS and CASIA-A datasets. The results show that the Siamese LSTM outperforms the gait feature-based approaches on the MARS dataset by 20% and 55% on the CASIA-A dataset. The results show that feature-based gait analysis using 2D and 3D pose estimators is premature. As a future study, we suggest developing large-scale human gait datasets and designing accurate 2D and 3D joint position estimators specifically for gait patterns. We expect that the current comparative study and the future work could contribute to rehabilitation study, forensic gait analysis and early detection of neurological disorders.

Improved Chinese Sentence Semantic Similarity Calculation Method Based on Multi-Feature Fusion

In this paper, an improved long short-term memory (LSTM)-based deep neural network structure is proposed for learning variable-length Chinese sentence semantic similarities. Siamese LSTM, a sequence-insensitive deep neural network model, has a limited ability to capture the semantics of natural language because it has difficulty explaining semantic differences based on the differences in syntactic structures or word order in a sentence. Therefore, the proposed model integrates the syntactic component features of the words in the sentence into a word vector representation layer to express the syntactic structure information of the sentence and the interdependence between words. Moreover, a relative position embedding layer is introduced into the model, and the relative position of the words in the sentence is mapped to a high-dimensional space to capture the local position information of the words. With this model, a parallel structure is used to map two sentences into the same high-dimensional space to obtain a fixed-length sentence vector representation. After aggregation, the sentence similarity is computed in the output layer. Experiments with Chinese sentences show that the model can achieve good results in the calculation of the semantic similarity.

Key Intrinsic Connectivity Networks for Individual Identification With Siamese Long Short-Term Memory.

In functional magnetic resonance imaging (fMRI) analysis, many studies have been conducted on inter-subject variability as well as intra-subject reproducibility. These studies indicate that fMRI could have unique characteristics for individuals. In this study, we hypothesized that the dynamic information during 1 min of fMRI was unique and repetitive enough for each subject, so we applied long short-term memory (LSTM) using initial time points of dynamic resting-state fMRI for individual identification. Siamese network is used to obtain robust individual identification performance without additional learning on a new dataset. In particular, by adding a new structure called region of interest–wise average pooling (RAP), individual identification performance could be improved, and key intrinsic connectivity networks (ICNs) for individual identification were also identified. The average performance of individual identification was 97.88% using the test dataset in eightfold cross-validation analysis. Through the visualization of features learned by Siamese LSTM with RAP, ICNs spanning the parietal region were observed as the key ICNs in identifying individuals. These results suggest the key ICNs in fMRI could represent individual uniqueness.

Development and Grasp Stability Estimation of Sensorized Soft Robotic Hand.

This paper introduces the development of an anthropomorphic soft robotic hand integrated with multiple flexible force sensors in the fingers. By leveraging on the integrated force sensing mechanism, grip state estimation networks have been developed. The robotic hand was tasked to hold the given object on the table for 1.5 s and lift it up within 1 s. The object manipulation experiment of grasping and lifting the given objects were conducted with various pneumatic pressure (50, 80, and 120 kPa). Learning networks were developed to estimate occurrence of object instability and slippage due to acceleration of the robot or insufficient grasp strength. Hence the grip state estimation network can potentially feedback object stability status to the pneumatic control system. This would allow the pneumatic system to use suitable pneumatic pressure to efficiently handle different objects, i.e., lower pneumatic pressure (50 kPa) for lightweight objects which do not require high grasping strength. The learning process of the soft hand is made challenging by curating a diverse selection of daily objects, some of which displays dynamic change in shape upon grasping. To address the cost of collecting extensive training datasets, we adopted one-shot learning (OSL) technique with a long short-term memory (LSTM) recurrent neural network. OSL aims to allow the networks to learn based on limited training data. It also promotes the scalability of the network to accommodate more grasping objects in the future. Three types of LSTM-based networks have been developed and their performance has been evaluated in this study. Among the three LSTM networks, triplet network achieved overall stability estimation accuracy at 89.96%, followed by LSTM network with 88.00% and Siamese LSTM network with 85.16%.

A New Online Plagiarism Detection System based on Deep Learning

The Plagiarism is an increasingly widespread and growing problem in the academic field. Several plagiarism techniques are used by fraudsters, ranging from a simple synonym replacement, sentence structure modification, to more complex method involving several types of transformation. Human based plagiarism detection is difficult, not accurate, and time-consuming process. In this paper we propose a plagiarism detection framework based on three deep learning models: Doc2vec, Siamese Long Short-term Memory (SLSTM) and Convolutional Neural Network (CNN). Our system uses three layers: Preprocessing Layer including word embedding, Learning Layers and Detection Layer. To evaluate our system, we carried out a study on plagiarism detection tools from the academic field and make a comparison based on a set of features. Compared to other works, our approach performs a good accuracy of 98.33 % and can detect different types of plagiarism, enables to specify another dataset and supports to compare the document from an internet search.

Learning to Match Clothing From Textual Feature-Based Compatible Relationships

This paper presents a new framework for matching clothes by considering item in-between compatibility. In contrast to the use of visual features of clothing items, we only utilized their textual descriptions, i.e., title sentences, to constitute the basic features. Specifically, a longshort-term memory (LSTM) network was used for feature embeddings of title sentences. Given item pairs of queries and candidates, their feature embeddings achieved by Siamese LSTMs were integrated into style-compatible space characterized by a compatibility matrix. Our framework is examined on three large-scaled clothing item sets collected from Amazon, Taobao, and Polyvore, respectively. Experiments confirm the efficacy of our approach compared with several baseline methods.