Sentence Similarity Research Articles

A dataset for plain language adaptation of biomedical abstracts

Though exponentially growing health-related literature has been made available to a broad audience online, the language of scientific articles can be difficult for the general public to understand. Therefore, adapting this expert-level language into plain language versions is necessary for the public to reliably comprehend the vast health-related literature. Deep Learning algorithms for automatic adaptation are a possible solution; however, gold standard datasets are needed for proper evaluation. Proposed datasets thus far consist of either pairs of comparable professional- and general public-facing documents or pairs of semantically similar sentences mined from such documents. This leads to a trade-off between imperfect alignments and small test sets. To address this issue, we created the Plain Language Adaptation of Biomedical s dataset. This dataset is the first manually adapted dataset that is both document- and sentence-aligned. The dataset contains 750 adapted abstracts, totaling 7643 sentence pairs. Along with describing the dataset, we benchmark automatic adaptation on the dataset with state-of-the-art Deep Learning approaches, setting baselines for future research.

Philippine Court Case Summarizer using Latent Semantic Analysis

Artificial Intelligence in the field of Law has introduced a lot of innovations to aid legal experts, and one of those are summarizing systems. Text summarization is the process of condensing text into a compact summary highlighting the important parts of the original text. As years pass by, numerous text summarization systems are being developed, and improvements keep getting made. These advancements have been helping humans in easing labor work with summarizing lengthy documents. With this, the researchers would like to propose a tool that will summarize legal documents, specifically cases obtained from the Philippine Supreme Court website. This includes extracting meanings and analyzation of words, and similarity of sentences from the original legal document obtained from the website of the Philippine Supreme Court. The tool will take the content of the original document, analyze the words, structure, and sentences according to context, and finally produce the summary. The proposed tool will be called SUMMIT, from the words “Summarize It”. The method to be used to summarize the documents using the tool is called Latent Semantic Analysis (LSA). The researchers will use Python as their programming language in developing the summarizer tool.

A Novel Approach: Tokenization Framework based on Sentence Structure in Indonesian Language

This study proposes a new approach in the sentence tokenization process. Sentence tokenization, which is known so far, is the process of breaking sentences based on spaces as separators. Space-based sentence tokenization only generates single word tokens. In sentences consisting of five words, tokenization will produce five tokens, one word each. Each word is a token. This process ignores the loss of the original meaning of the separated words. Our proposed tokenization framework can generate one-word tokens and multi-word tokens at the same time. The process is carried out by extracting the sentence structure to obtain sentence elements. Each sentence element is a token. There are five sentence elements that is Subject, Predicate, Object, Complement and Adverbs. We extract sentence structures using deep learning methods, where models are built by training the datasets that have been prepared before. The training results are quite good with an F1 score of 0.7 and it is still possible to improve. Sentence similarity is the topic for measuring the performance of one-word tokens compared to multi-word tokens. In this case the multiword token has better accuracy. This framework was created using the Indonesian language but can also use other languages with dataset adjustments.

Investigation on Picture Books Reading Education in Primary School <br/>—Taking H Province for Example

Picture book reading facilitates the acquisition of reading literacy. How picture books are used in primary school education and whether picture book reading foster students’ learning ability are the two questions to be discussed in the passage. According to the investigation, lack of time, difficult in choosing picture books and the renovation of teaching methods are the main problems exist. In order to promote the reading literacy: teachers should categorize picture books or stories with the same or similar topics, grammar, words and sentence structure in their ordinary teaching; Teachers or parents should know a general form of asking questions to enhance the effect of students’ in picture book reading and improve students’ reading literacy.

On Robustness and Sensitivity of a Neural Language Model: A Case Study on Italian L1 Learner Errors

In this paper, we propose a comprehensive linguistic study aimed at assessing the implicit behavior of one of the most prominent Neural Language Models (NLM) based on Transformer architectures, BERT Devlin et al., when dealing with a particular source of noisy data, namely essays written by L1 Italian learners containing a variety of errors targeting grammar, orthography and lexicon. Differently from previous works, we focus on the pre-training stage and we devise two complementary evaluation tasks aimed at assessing the impact of errors on sentence-level inner representations in terms of semantic robustness and linguistic sensitivity. While the first evaluation perspective is meant to probe the model's ability to encode the semantic similarity between sentences also in the presence of errors, the second type of probing task evaluates the influence of errors on BERT's implicit knowledge of a set of raw and morpho-syntactic properties of a sentence. Our experiments show that BERT's ability to compute sentence similarity and to correctly encode multi-leveled linguistic information of a sentence are differently modulated by the category of errors and that the error hierarchies in terms of robustness and sensitivity change across layer-wise representations.

Heterogeneous Semantic and Bit Communications: A Semi-NOMA Scheme

Multiple access (MA) design is investigated to facilitate the coexistence of the emerging semantic transmission and the conventional bit-based transmission in future networks. The <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">semantic rate</i> is adopted for measuring the performance of the semantic transmission. However, a key challenge is that there is no closed-form expression for a key parameter, namely the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">semantic similarity</i> , which characterizes the sentence similarity between an original sentence and the corresponding recovered sentence. To overcome this challenge, we propose a data regression method, where the semantic similarity is approximated by a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">generalized logistic function</i> . Using the obtained tractable function, we propose a heterogeneous semantic and bit communication framework, where an access point simultaneously sends the semantic and bit streams to one semantics-interested user (S-user) and one bit-interested user (B-user). To realize this heterogeneous semantic and bit transmission in multi-user networks, three MA schemes are proposed, namely orthogonal multiple access (OMA), non-orthogonal multiple access (NOMA), and semi-NOMA. More specifically, the bit stream in semi-NOMA is split into two streams, one is transmitted with the semantic stream over the shared frequency sub-band and the other is transmitted over the separate orthogonal frequency sub-band. To study the fundamental performance limits of the three proposed MA schemes, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">semantic-versus-bit (SvB) rate region</i> and the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">power region</i> are defined. An optimal resource allocation procedure is then derived for characterizing the boundary of the SvB rate region and the power region achieved by each MA scheme. The structures of the derived solutions demonstrate that semi-NOMA is superior to both NOMA and OMA given its highly flexible transmission policy. Our numerical results: 1) confirm that the proposed semi-NOMA is the optimal MA scheme as compared to OMA and NOMA even under the symmetric channel case, and 2) reveal that the superiority of semi-NOMA is more prominent when the channel condition of the S-user is better than that of the B-user.

Question-Answering Pair Matching Based on Question Classification and Ensemble Sentence Embedding

Question-answering (QA) models find answers to a given question. The necessity of automatically finding answers is increasing because it is very important and challenging from the large-scale QA data sets. In this paper, we deal with the QA pair matching approach in QA models, which finds the most relevant question and its recommended answer for a given question. Existing studies for the approach performed on the entire dataset or datasets within a category that the question writer manually specifies. In contrast, we aim to automatically find the category to which the question belongs by employing the text classification model and to find the answer corresponding to the question within the category. Due to the text classification model, we can effectively reduce the search space for finding the answers to a given question. Therefore, the proposed model improves the accuracy of the QA matching model and significantly reduces the model inference time. Furthermore, to improve the performance of finding similar sentences in each category, we present an ensemble embedding model for sentences, improving the performance compared to the individual embedding models. Using real-world QA data sets, we evaluate the performance of the proposed QA matching model. As a result, the accuracy of our final ensemble embedding model based on the text classification model is 81.18%, which outperforms the existing models by 9.81%∼14.16% point. Moreover, in terms of the model inference speed, our model is faster than the existing models by 2.61∼5.07 times due to the effective reduction of search spaces by the text classification model.

Joint span and token framework for few-shot named entity recognition

Few-shot Named Entity Recognition (NER) is a challenging task that involves identifying new entity types using a limited number of labeled instances for training. Currently, the majority of Few-shot NER methods are based on span, which pay more attention to the boundary information of the spans as candidate entities and the entity-level information. However, these methods often overlook token-level semantic information, which can limit their effectiveness. To address this issue, we propose a novel Joint Span and Token (JST) framework that integrates both the boundary information of an entity and the semantic information of each token that comprises an entity. The JST framework employs span features to extract the boundary features of the entity and token features to extract the semantic features of each token. Additionally, to reduce the negative impact of the Other class, we introduce a method to separate named entities from the Other class in semantic space, which helps to improve the distinction between entities and the Other class. In addition, we used GPT to do data augmentation on the support sentences, generating similar sentences to the original ones. These sentences increase the diversity of the sample and the reliability of our model. Our experimental results on the Few-NERD11https://ningding97.github.io/fewnerd/. and SNIPS22https://github.com/AtmaHou/FewShotTagging. datasets demonstrate that our model outperforms existing methods in terms of performance.

A Writer in the Making - A Comparative Analysis of Ellen G. White’s Writings on the Resurrection of Christ -

재림교회 창시자 중 한 사람인 엘렌 지 화잇은 많은 저술을 남겼다. 그의 글은 신학자들에 의해 광범위하게 연구되었지만 문학적 또는 언어학적 관점에서 조사한 연구는 상대적으로 적다. 화잇은 작문에 대한 정규 수업을 거의 받지 못했지만 그의 글은 지금까지 수많은 사람에게 읽히고 있다. 그러므로 그녀가 작가로서어떻게 성숙했는지 파악하는 연구는 가치가 있다. 특별히 이 연구는 번역본을읽었기 때문에 그의 텍스트가 보여주는 다양한 글쓰기 스타일에 익숙하지 않은해외 독자들에게 흥미있는 연구가 될 것이다. 소수의 기존 연구는 대쟁투 총서특히 화잇이 60대였을 때 출판한 ≷시대의 소망≸과 같은 대표적인 작품을 중심으로 그의 글쓰기 스타일을 비교 조사했다. 이러한 연구는 그의 글의 독창성에 의문을 제기한 회의론자들을 반박하는 것에 주로 초점을 맞췄다. 현재까지 화잇의 초기저작물을 ≷시대의 소망≸이 출판되기 수십 년 전에 작성된 저작물과 문체면에서어떻게 다른지 비교 조사한 연구는 찾아보기 힘들다. 이 간극을 메우기 위해 이논문은 20년 간격으로 저술된 두 권의 책인 ≷영적선물 I≸과 ≷예언의 신 III≸에서예수의 부활을 다루는 두 발췌문을 비교 분석한다. 이 연구는 작가로서의 화잇의성장을 조명하기 위해 문단의 구성, 언어, 어조, 문체 및 수사적 기교에 초점을맞춘다. 이 연구는 화잇이 두 텍스트에서 유사한 문장을 재사용하지만 이후 텍스트에서 내용, 구성, 문법 및 문구를 다양하게 변경하여 보다 향상된 문어체 스타일과 어조를 썼으며, 이러한 성숙한 스타일은 ≷시대의 소망≸에 대한 이전 연구에서관찰한바와 일치함을 보여준다.Ellen G. White, one of the founders of the Seventh-day Adventist (SDA) Church, was a prolific writer. While her writings have been extensively studied by theologians, relatively few studies have examined her writings from a literary or linguistic perspective. Despite her lack of formal education and training in writing, her writings continue to draw and inspire readers. Therefore, it would be worthwhile to examine White’s texts to understand how she developed as a writer. This can be a particularly illuminating task for international readers who have mainly read translations of her books, and therefore, are not as familiar with the different writing styles that her various texts exhibit. A small number of existing studies have comparatively examined White’s writing styles at different time periods by focusing on some of her most representative works, such as the Conflict of the Ages series, and The Desire of the Ages, in particular, which was published when White was in her 60s. These studies largely focused on addressing skeptics who questioned the originality of her writing. To date, there is a dearth of studies that examined how White’s earlier writings compare to those written in the decades prior to the publication of The Desire of Ages. To help fill this gap, this paper comparatively analyzes two excerpts that deal with Jesus’ resurrection from White’s two different books—Spiritual Gifts I, and The Spirit of Prophecy III―which were written two decades apart. The current study focuses on the organization, language, tone, style, and rhetorical moves in order to shed further light on her development as a writer. The study shows that White reuses similar sentences in the two texts but makes various changes to the content, organization, grammar, and wording in her later text, which in turn create a more formal style and tone, and that the improved style aligns with the observations made by previous studies of The Desire of Ages.

Analisis Hukum Terhadap Pertimbangan Hakim Atas Vonis Nihil Kepada Pelaku Tindak Pidana Korupsi

This study aims to determine and analyze the verdict of zero in the corruption case committed by the defendant on behalf of Heru Hidayat which was reported in the corruption case at PT. Asuransi Jiwasraya which cost the state Rp. 16 trillion and the defendant was sentenced to life imprisonment, but for the Corruption Crime Case at PT. ASABRI caused a loss to the state of Rp. 22.78 trillion, but the defendant and the judge handed down a zero sentence verdict, of course this greatly disturbed the justice of the people in Indonesia. This research method uses a type of normative juridical research, namely research conducted by means of a literature study and a case approach. After the data is collected, a qualitative analysis is carried out. The results of this study are to explain that a zero sentence is regulated in Article 67 of the Criminal Code which states that if someone has been sentenced to death or life imprisonment, then no other punishment may be imposed except for the revocation of certain rights. However, these provisions limit the possibility of people who commit various criminal acts who are then tried either at the same time or tried separately for a total of more than 20 years in prison. Based on Article 272 of the Criminal Procedure Code, the convict is sentenced to imprisonment or confinement and then sentenced to a similar sentence before serving the sentence previously imposed, then the sentence is carried out successively starting with the sentence that was imposed first

Bertalign: Improved word embedding-based sentence alignment for Chinese–English parallel corpora of literary texts

Abstract Bertalign is designed to improve sentence alignment accuracy for Chinese–English parallel corpora of literary texts. Aligning bilingual literary texts is not trivial, since most of the translation is interpretative and not based on 1-to-1 mappings between source and target sentences. Existing alignment methods highlight 1-to-1 links while having difficulty coping with 1-to-many and many-to-many alignments that are common in literary texts. To overcome the weaknesses of current approaches, we propose a novel two-step algorithm for bilingual sentence alignment. The first step finds the optimal paths for 1-to-1 alignments based on the top-k most semantically similar target sentences for each source sentence using the bidirectional encoder representations from transformer-based cross-lingual word embeddings. The second step relies on search paths found in the previous step to recover all valid alignments with more than one sentence on each side of the bilingual text. A comprehensive experiment was conducted on a newly built Chinese–English literary parallel corpus and a large-scale publicly available bilingual corpus of the Bible to compare the performance of Bertalign with five baseline systems: Gale-Church, Hunalign, Bleualign, Bleurtalign, and Vecalign. The results show that Bertalign achieves the highest accuracy in terms of F1 score on the two evaluation datasets than previous methods.

REWRITING IDENTIFICATION TECHNOLOGY FOR TEXT CONTENT BASED ON MACHINE LEARNING METHODS

Context. Paraphrased textual content or rewriting is one of the difficult problems of detecting academic plagiarism. Most plagiarism detection systems are designed to detect common words, sequences of linguistic units, and minor changes, but are unable to detect significant semantic and structural changes. Therefore, most cases of plagiarism using paraphrasing remain unnoticed.&#x0D; Objective of the study is to develop a technology for detecting paraphrasing in text based on a classification model and machine learning methods through the use of Siamese neural network based on recurrent and Transformer type – RoBERTa to analyze the level of similarity of sentences of text content.&#x0D; Method. For this study, the following semantic similarity metrics or indicators were chosen as features: Jacquard coefficient for shared N-grams, cosine distance between vector representations of sentences, Word Mover’s Distance, distances according to WordNet dictionaries, prediction of two ML models: Siamese neural network based on recurrent and Transformer type - RoBERTa.&#x0D; Results. An intelligent system for detecting paraphrasing in text based on a classification model and machine learning methods has been developed. The developed system uses the principle of model stacking and feature engineering. Additional features indicate the semantic affiliation of the sentences or the normalized number of common N-grams. An additional fine-tuned RoBERTa neural network (with additional fully connected layers) is less sensitive to pairs of sentences that are not paraphrases of each other. This specificity of the model may contribute to incorrect accusations of plagiarism or incorrect association of user-generated content. Additional features increase both the overall classification accuracy and the model’s sensitivity to pairs of sentences that are not paraphrases of each other.&#x0D; Conclusions. The created model shows excellent classification results on PAWS test data: precision – 93%, recall – 92%, F1score – 92%, accuracy – 92%. The results of the study showed that Transformer-type NNs can be successfully applied to detect paraphrasing in a pair of texts with fairly high accuracy without the need for additional feature generation.

Informing the development of an outcome set and banks of items to measure mobility among individuals with acquired brain injury using natural language processing.

The sheer number of measures evaluating mobility and inconsistencies in terminology make it challenging to extract potential core domains and items. Automating a portion of the data synthesis would allow us to cover a much larger volume of studies and databases in a smaller fraction of the time compared to the usual process. Thus, the objective of this study was to identify a comprehensive outcome set and develop preliminary banks of items of mobility among individuals with acquired brain injury (ABI) using Natural Language Processing (NLP). An umbrella review of 47 reviews evaluating the content of mobility measures among individuals with ABI was conducted. A search was performed on 5 databases between 2000 and 2020. Two independent reviewers retrieved copies of the measures and extracted mobility domains and items. A pre-trained BERT model (state-of-the-art model for NLP) provided vector representations for each sentence. Using the International Classification of Functioning, Disability, and Health Framework (ICF) ontology as a guide for clustering, a k-means algorithm was used to retrieve clusters of similar sentences from their embeddings. The resulting embedding clusters were evaluated using the Silhouette score and fine-tuned according to expert input. The study identified 246 mobility measures, including 474 domains and 2109 items. Encoding the clusters using the ICF ontology and expert knowledge helped in regrouping the items in a way that is more closely related to mobility terminology. Our best results identified banks of items that were used to create a 24 comprehensive outcome sets of mobility, including Upper Extremity Mobility, Emotional Function, Balance, Motor Control, Self-care, Social Life and Relationships, Cognition, Walking, Postural Transition, Recreation, and Leisure Activities, Activities of Daily Living, Physical Functioning, Communication, Work/Study, Climbing, Sensory Functions, General Health, Fatigue, Functional Independence, Pain, Alcohol and Drugs Use, Transportation, Sleeping, and Finances. The banks of items of mobility domains represent a first step toward establishing a comprehensive outcome set and a common language of mobility to develop the ontology. It enables researchers and healthcare professionals to begin exposing the content of mobility measures as a way to assess mobility comprehensively.

FEATURE ENGINEERING WITH SENTENCE SIMILARITY USING THE LONGEST COMMON SUBSEQUENCE FOR EMAIL CLASSIFICATION

Feature selection plays a prominent role in email classification since selecting the most relevant features enhances the accuracy and performance of the learning classifier. Due to the exponential increase rate in the usage of emails, the classification of such emails posed a fitting problem. Therefore, there is a requirement for a proper classification system. Such an email classification system requires an efficient feature selection method for the accurate classification of the most relevant features. This paper proposes a novel feature selection method for sentence similarity using the longest common subsequence for email classification. The proposed feature selection method works in two main phases: First, it builds the longest common subsequence vector of features by comparing each email with all other emails in the dataset. Later, a template is constructed for each class using the closest features of emails of a particular class. Further, email classification is tested for unseen emails using these templates. The performance of the proposed method is compared with traditional feature selection methods such as TF-IDF, Information Gain, Chi-square, and semantic approach. The experimental results showed that the proposed method performed well with 96.61% accuracy.

Validation of a Thai artificial chatmate designed for cheering up the elderly during the COVID-19 pandemic

Background: The COVID-19 pandemic severely affected populations of all age groups. The elderly are a high-risk group and are highly vulnerable to COVID-19. Assistive software chatbots can enhance the mental health status of the elderly by providing support and companionship. The objective of this study was to validate a Thai artificial chatmate for the elderly during the COVID-19 pandemic and floods. Methods: Chatbot design includes the establishment of a dataset and emotional word vectors in which data consisting of emotional sentences were converted into the word vector form using a pre-trained word2vec model. A word vector was then input into a convolutional neural network (CNN) and trained until the model converges using sentence embedding and similarity word segmentation. Sentence vectors were generated by averaging each word vector using an averaged vector method. For approximate similarity matching, the Annoy library was used to create the indices in tree sorting. Data were collected from 22 elderly and assessed by the Post-Study System Usability Questionnaire (PSSUQ). Results: The study revealed that 72.73% of the respondents found the chatbot easy to learn and use, 63.64% of the respondents found the chatbot can autonomously determine the next course of action, and 59.09% of the respondents believed that troubleshooting guidelines were provided for overcoming errors. The accuracy of the chatbot providing a reasonable response is 56.20±13.99%. Conclusions: Most users were satisfied with the chatbot system. The proposed chatbot provided considerable essential insights into the development of assistance systems for the elderly during the coronavirus pandemic (COVID-19) and during the period of national disasters. The model can be expanded to other applications in the future.

A Pragmatic and Systemic Approach to Advance Research in Health Policy and Management Comment on "Insights Gained From a Re-analysis of Five Improvement Cases in Healthcare Integrating System Dynamics Into Action Research".

"Healthcare is complex" - or similar sentences - is a statement that introduces a wide number of scientific articles dealing with health policy and management issues. We all agree that healthcare is complex, but most studies, although using this kind of sentence to introduce their background, do little to effectively deal with such complexity in their analyses. Holmström et al proposed a methodological approach to tackle healthcare complexity by integrating system dynamics (SD) into action research (AR). This commentary highlights three touch points that makes the combination of AR and SD feasible, namely the epistemological ground, the use of experimentation and the collaborative approach. The proposed approach addresses some of the key sources of the complexity characterizing healthcare settings.

A reproducible experimental survey on biomedical sentence similarity: A string-based method sets the state of the art.

This registered report introduces the largest, and for the first time, reproducible experimental survey on biomedical sentence similarity with the following aims: (1) to elucidate the state of the art of the problem; (2) to solve some reproducibility problems preventing the evaluation of most current methods; (3) to evaluate several unexplored sentence similarity methods; (4) to evaluate for the first time an unexplored benchmark, called Corpus-Transcriptional-Regulation (CTR); (5) to carry out a study on the impact of the pre-processing stages and Named Entity Recognition (NER) tools on the performance of the sentence similarity methods; and finally, (6) to bridge the lack of software and data reproducibility resources for methods and experiments in this line of research. Our reproducible experimental survey is based on a single software platform, which is provided with a detailed reproducibility protocol and dataset as supplementary material to allow the exact replication of all our experiments and results. In addition, we introduce a new aggregated string-based sentence similarity method, called LiBlock, together with eight variants of current ontology-based methods, and a new pre-trained word embedding model trained on the full-text articles in the PMC-BioC corpus. Our experiments show that our novel string-based measure establishes the new state of the art in sentence similarity analysis in the biomedical domain and significantly outperforms all the methods evaluated herein, with the only exception of one ontology-based method. Likewise, our experiments confirm that the pre-processing stages, and the choice of the NER tool for ontology-based methods, have a very significant impact on the performance of the sentence similarity methods. We also detail some drawbacks and limitations of current methods, and highlight the need to refine the current benchmarks. Finally, a notable finding is that our new string-based method significantly outperforms all state-of-the-art Machine Learning (ML) models evaluated herein.

Task-dependent Optimal Weight Combinations for Static Embeddings

A variety of NLP applications use word2vec skip-gram, GloVe, and fastText word embeddings. These models learn two sets of embedding vectors, but most practitioners use only one of them, or alternately an unweighted sum of both. This is the first study to systematically explore a range of linear combinations between the first and second embedding sets. We evaluate these combinations on a set of six NLP benchmarks including IR, POS-tagging, and sentence similarity. We show that the default embedding combinations are often suboptimal and demonstrate 1.0-8.0% improvements. Notably, GloVe’s default unweighted sum is its least effective combination across tasks. We provide a theoretical basis for weighting one set of embeddings more than the other according to the algorithm and task. We apply our findings to improve accuracy in applications of cross-lingual alignment and navigational knowledge by up to 15.2%.

PERANAN PERANGKAT TURNITIN DALAM MENDORONG KARYA ILMIAH BERKUALITAS

The purpose of scientific writing is to explain the role of the Turnitin tool to detect plagiarism in a scientific work and also to encourage quality scientific work. To suppress plagiarism and encourage quality scientific work, it is necessary to have a detailed and intense mastery of scientific work material so as to produce scientific writings that produce writings that have high originality. To suppress plagiarism in scientific writing, you can also use a paraphrase strategy and process words and sentences manually. Although the Turnitin tool is currently the best tool for detecting plagiarism, it still has a weakness, namely that it only detects the similarity or similarity of words and sentences. For this reason, there needs to be collaboration between students, lecturers and universities in suppressing thesis plagiarism and utilizing the role of the Turnitin tool in encouraging quality scientific works or thesis..

CURIOUS: Efficient Neural Architecture Search Based on a Performance Predictor and Evolutionary Search

Neural networks (NNs) have been successfully deployed in various applications of artificial intelligence. However, architectural design of NNs is still a challenging problem. This is due to the need to navigate a search space based on a large number of hyperparameters. This forces the search space of possible architectures to grow exponentially. Using a trial-and-error design approach is very time consuming and leads to suboptimal architectures. In addition, approaches, such as neural architecture search based on reinforcement learning and differentiable gradient-based architecture search, often incur huge computational costs or significant memory requirements. To address these challenges, we propose the CURIOUS NN synthesis methodology. It uses a performance predictor to efficiently navigate the architectural search space with an evolutionary search process. The predictor is built using quasi Monte-Carlo sampling, boosted decision tree regression, and an intelligent iterative sampling method. It is designed to be sample efficient. CURIOUS starts from a base architecture and explores the architectural search space to obtain a variant of the base architecture with the highest performance. This search framework is general and covers all important NN architecture types, e.g., feedforward NNs (FFNNs), convolutional NNs (CNNs), recurrent NNs (RNNs), and transformers. We evaluate the performance of CURIOUS on various datasets and base architectures. Through these experiments, we demonstrate significant performance improvements over the baseline architectures. For the MNIST dataset, our CNN architecture achieves an error rate of 0.66%, with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$8.6\times $ </tex-math></inline-formula> fewer parameters compared to the LeNet-5 baseline. For the CIFAR-10 dataset, we use the ResNet architectures and residual networks with Shake-Shake regularization as the baselines. Our synthesized ResNet-18 has a 2.52% accuracy improvement over the original ResNet-18, 1.74% over ResNet-101, and 0.16% over ResNet-1001, while requiring comparable number of parameters and floating-point operations to the original ResNet-18. This result shows that instead of just increasing the number of layers to increase accuracy, an alternative is to use a better NN architecture with a small number of layers. In addition, CURIOUS achieves an error rate of just 2.69% with a variant of the residual architecture with Shake-Shake regularization. We also use the set of optimized hyperparameters found for ResNet-18 on the CIFAR-10 dataset to train and evaluate the model on the ImageNet dataset, and show 3.43% (1.83%) improvement in the top-1 (top-5) error rate compared to the original ResNet-18 model. CURIOUS also obtains the highest accuracy for various other FFNNs that are geared toward edge devices and IoT sensors. In addition, we use CURIOUS to search for deep RNN architectures for the SICK dataset for sentence similarity evaluation. It achieves a mean-squared error of only 0.2060, improving upon the base network performance, without the need to stack multiple long short-term memories. We also use CURIOUS to search for a better NN classifier for the sentiment analysis task on the Stanford sentiment treebank dataset using a pretrained BERT model and again demonstrate improvements in performance.