Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558947
Tuğba Bekman
Product piracy, which remains a major problem in world trade, affects every actor in the supply chain. In order to combat product piracy, legal and administrative measures are taken as well as measures from production to sale, while product identification and tracking play an important role. Labeling technologies such as RFID / NFC tags and QR codes are often used for this. However, data stored in RFID / NFC tags and QR codes can be manipulated. This weakness of the technologies mentioned could be remedied with blockchain technology. Based on this idea, this thesis examines which solutions can support the features of blockchain technology in the fight against piracy. In the first part of this work the reader is given a certain framework for product piracy, in the second part current measures of manufacturers against product piracy are explained. The third chapter explains the basic functions of blockchain technology. The last part describes different usage scenarios of blockchain technology to prevent product piracy. By comparing these scenarios, common and different characteristics of blockchain projects are identified.
{"title":"Use of Blockchain Technology to Fight Trade in Counterfeit Goods","authors":"Tuğba Bekman","doi":"10.1109/UBMK52708.2021.9558947","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558947","url":null,"abstract":"Product piracy, which remains a major problem in world trade, affects every actor in the supply chain. In order to combat product piracy, legal and administrative measures are taken as well as measures from production to sale, while product identification and tracking play an important role. Labeling technologies such as RFID / NFC tags and QR codes are often used for this. However, data stored in RFID / NFC tags and QR codes can be manipulated. This weakness of the technologies mentioned could be remedied with blockchain technology. Based on this idea, this thesis examines which solutions can support the features of blockchain technology in the fight against piracy. In the first part of this work the reader is given a certain framework for product piracy, in the second part current measures of manufacturers against product piracy are explained. The third chapter explains the basic functions of blockchain technology. The last part describes different usage scenarios of blockchain technology to prevent product piracy. By comparing these scenarios, common and different characteristics of blockchain projects are identified.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132240516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558963
Ahmet Semih Uçan, Fatih Mustafa Buçak, Mehmet Ali Han Tutuk, Halis İbrahim Aydin, Ertuğrul Semiz, Şerif Bahtiyar
Deep learning is widely used to create artificial contents on the Internet. Similarly, it is also used to detect fake contents. Fake frames created and integrated with deep learning algorithms are known as deepfake. Recently, malicious users tend to use deepfake to manipulate genuine contents to carry out variety of attacks. Video conferencing applications has been a significant target of the malicious users since the beginning of Covid-19 pandemic who use deepfake models to create fake virtual identities in online video conferences. We propose a lightweight deepfake detection model that may be integrated with video conference applications to detect fake faces. Experimental analyses show that the proposed model provides acceptable accuracy to detect fake images on video conferences.
{"title":"Deepfake and Security of Video Conferences","authors":"Ahmet Semih Uçan, Fatih Mustafa Buçak, Mehmet Ali Han Tutuk, Halis İbrahim Aydin, Ertuğrul Semiz, Şerif Bahtiyar","doi":"10.1109/UBMK52708.2021.9558963","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558963","url":null,"abstract":"Deep learning is widely used to create artificial contents on the Internet. Similarly, it is also used to detect fake contents. Fake frames created and integrated with deep learning algorithms are known as deepfake. Recently, malicious users tend to use deepfake to manipulate genuine contents to carry out variety of attacks. Video conferencing applications has been a significant target of the malicious users since the beginning of Covid-19 pandemic who use deepfake models to create fake virtual identities in online video conferences. We propose a lightweight deepfake detection model that may be integrated with video conference applications to detect fake faces. Experimental analyses show that the proposed model provides acceptable accuracy to detect fake images on video conferences.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132308998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558891
Hilal Tekgöz, Halil Ibrahim Celenli, S. İ. Omurca
Natural Language Processing has become an important issue with the rapid increase in textual data in the health sector recently. Especially with the effect of COVID-19, easy and fast analysis of health data is important for research. Traditional text representations such as BoW (bag of words), TF-IDF (term frequency-inverse document frequency), and modern word representation methods such as FastText and BERT are used to represent words. The BERT models are provided high performance recently. The BERT models are divided into pre-trained and fine-tuned BERT models. In order to get good results in the field of health, BioBERT models are obtained by fine-tuning the basic BERT models with datasets containing biomedical articles. In this study, semantic similarities in datasets are evaluated by the Pearson correlation method by using BoW, TF-IDF, FastText, BERT, and BioBERT models. As a result of the evaluations, it was observed that BioBERT models gave higher values compared to other models and methods used.
{"title":"Semantic Similarity Comparison of Word Representation Methods in the Field of Health","authors":"Hilal Tekgöz, Halil Ibrahim Celenli, S. İ. Omurca","doi":"10.1109/UBMK52708.2021.9558891","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558891","url":null,"abstract":"Natural Language Processing has become an important issue with the rapid increase in textual data in the health sector recently. Especially with the effect of COVID-19, easy and fast analysis of health data is important for research. Traditional text representations such as BoW (bag of words), TF-IDF (term frequency-inverse document frequency), and modern word representation methods such as FastText and BERT are used to represent words. The BERT models are provided high performance recently. The BERT models are divided into pre-trained and fine-tuned BERT models. In order to get good results in the field of health, BioBERT models are obtained by fine-tuning the basic BERT models with datasets containing biomedical articles. In this study, semantic similarities in datasets are evaluated by the Pearson correlation method by using BoW, TF-IDF, FastText, BERT, and BioBERT models. As a result of the evaluations, it was observed that BioBERT models gave higher values compared to other models and methods used.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114569681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558953
Zhela Jalal Rashid, M. F. Adak
Random test data generators are among the most widely used tools to generate input data for the tests. However, the data types and parameters have to be manually tailored into the tools and need to be updated manually once the source code or the test cases are changed. It is a costly process and takes a lot of time and effort to generate and update these data. Various test data generator tools are available, such as random test data generators, symbolic evaluators, and function minimization methods. In recent years some more advanced heuristic search techniques have been applied to software testing. In this study, we propose a model which automates the test data generation process. It significantly reduces the time required to generate the input data. At the same time, the data generated by our model outperforms the data generated randomly in terms of the accuracy and sensibility of the input data. It is based on the most widely used heuristic algorithm, the genetic algorithm (GA). We run the model on a sample class with six independent public methods of the different method signature, return type, and several arguments. It takes 5 seconds to generate ten possible inputs for each method with a mean, standard deviation of 0.15 and best candidate fitness average of 8.82, and means fitness of 9.79.
{"title":"Test Data Generation for Dynamic Unit Test in Java Language using Genetic Algorithm","authors":"Zhela Jalal Rashid, M. F. Adak","doi":"10.1109/UBMK52708.2021.9558953","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558953","url":null,"abstract":"Random test data generators are among the most widely used tools to generate input data for the tests. However, the data types and parameters have to be manually tailored into the tools and need to be updated manually once the source code or the test cases are changed. It is a costly process and takes a lot of time and effort to generate and update these data. Various test data generator tools are available, such as random test data generators, symbolic evaluators, and function minimization methods. In recent years some more advanced heuristic search techniques have been applied to software testing. In this study, we propose a model which automates the test data generation process. It significantly reduces the time required to generate the input data. At the same time, the data generated by our model outperforms the data generated randomly in terms of the accuracy and sensibility of the input data. It is based on the most widely used heuristic algorithm, the genetic algorithm (GA). We run the model on a sample class with six independent public methods of the different method signature, return type, and several arguments. It takes 5 seconds to generate ten possible inputs for each method with a mean, standard deviation of 0.15 and best candidate fitness average of 8.82, and means fitness of 9.79.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134345812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9559025
Serhat Celil, Selçuk Aslan, Sercan Demirci
Big data is a topic that is increasing in popularity day by day, and new techniques are being developed for the optimization processes performed on it. Harmony Search (HS) algorithm, inspired by music and harmonies, is an intuitive algorithm and has been used for the optimization of many problems. In this study, a new technique called source-linked HS algorithm (slinkHSA) focusing on big data optimization problems is presented. Experimental results were obtained with the slinkHS algorithm, results were compared with other popular metaheuristic algorithms and unmodified HS algorithm. The obtained results showed that the technique applied in the slinkHS algorithm adapted to the problem better, in this way better results could be obtained than other algorithms compared.
{"title":"A Novel Harmony Search Based Method for Noise Minimization on EEG Signals","authors":"Serhat Celil, Selçuk Aslan, Sercan Demirci","doi":"10.1109/UBMK52708.2021.9559025","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9559025","url":null,"abstract":"Big data is a topic that is increasing in popularity day by day, and new techniques are being developed for the optimization processes performed on it. Harmony Search (HS) algorithm, inspired by music and harmonies, is an intuitive algorithm and has been used for the optimization of many problems. In this study, a new technique called source-linked HS algorithm (slinkHSA) focusing on big data optimization problems is presented. Experimental results were obtained with the slinkHS algorithm, results were compared with other popular metaheuristic algorithms and unmodified HS algorithm. The obtained results showed that the technique applied in the slinkHS algorithm adapted to the problem better, in this way better results could be obtained than other algorithms compared.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134008443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As in all other fields, linguistics is accelerating the process of adapting to digital technologies. Consequently, it is important to process the traditional linguistic norms of natural language for computer programs and information systems. One such important task in NLP is tagging parts of speech. Part-of-speech tagging (abbreviation: (POS tagging or PoS tagging or POST) in Russian “частеречная разметка”) is a stage of automatic text processing, the function of it which is a series of words (forms) used in the text and it is to determine grammatical features. With this function, POS-tagging is one of the first steps in automatic text analysis. The article discusses the need for tagging parts of speech, tagging methods.
{"title":"Methods of Tagging Part of Speech of Uzbek Language","authors":"Abjalova Manzura Abdurashetona, Iskandarov Otabek Ismailovich","doi":"10.1109/UBMK52708.2021.9558900","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558900","url":null,"abstract":"As in all other fields, linguistics is accelerating the process of adapting to digital technologies. Consequently, it is important to process the traditional linguistic norms of natural language for computer programs and information systems. One such important task in NLP is tagging parts of speech. Part-of-speech tagging (abbreviation: (POS tagging or PoS tagging or POST) in Russian “частеречная разметка”) is a stage of automatic text processing, the function of it which is a series of words (forms) used in the text and it is to determine grammatical features. With this function, POS-tagging is one of the first steps in automatic text analysis. The article discusses the need for tagging parts of speech, tagging methods.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134110971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558881
A. Onay, Yasin Akın, Ali Kafalı, Erol Çıracı
Nowadays, air and water quality are a big problem in the cities because they affect human health deeply. To determine air and water quality, the monitoring system containing a wireless sensor node having several sensors has been developed with the internet of things (IoT) technology in this study. It regularly monitors various parameters such as temperature, humidity, bar pressure, wind direction, wind speed max, wind speed average, rain fall one hour, rain fall one day, $PM_{2.5}$ and PM10 for air quality monitoring and consumed water, waste water, incoming water, pH sensor, conductivity sensor, turbidity sensor, dissolved oxygen and temperature sensors for water quality monitoring. To improve smart city environments, air and water quality monitoring has to be common, present everywhere, and rapidly responsive. The monitoring system based on IoT is able to track air and water pollution in real time and transmit the information fast through a wide area network. This system can be integrated with innovative approaches like smart city. Therefore, the demand for a real time monitoring system will increase day by day.
{"title":"Real Time Air and Water Quality Monitoring based on Distributed Sensor Network","authors":"A. Onay, Yasin Akın, Ali Kafalı, Erol Çıracı","doi":"10.1109/UBMK52708.2021.9558881","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558881","url":null,"abstract":"Nowadays, air and water quality are a big problem in the cities because they affect human health deeply. To determine air and water quality, the monitoring system containing a wireless sensor node having several sensors has been developed with the internet of things (IoT) technology in this study. It regularly monitors various parameters such as temperature, humidity, bar pressure, wind direction, wind speed max, wind speed average, rain fall one hour, rain fall one day, $PM_{2.5}$ and PM10 for air quality monitoring and consumed water, waste water, incoming water, pH sensor, conductivity sensor, turbidity sensor, dissolved oxygen and temperature sensors for water quality monitoring. To improve smart city environments, air and water quality monitoring has to be common, present everywhere, and rapidly responsive. The monitoring system based on IoT is able to track air and water pollution in real time and transmit the information fast through a wide area network. This system can be integrated with innovative approaches like smart city. Therefore, the demand for a real time monitoring system will increase day by day.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133987217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558979
R. Rajabioun, Ali Özen Akyürek, E. Sezer
Electroencephalography (EEG) signals are normally susceptible to various artifacts and noises from different sources. In this paper, firstly the existence of artifacts will be identified on the recorded EEG signals and then the origin of the detected artifact will be determined among 7 different sources. Due to the nature of EEG signals, almost no specialist can determine artifact source through eye inspection. This paper introduces the utilization of 1-D Convolutional Neural Network (CNN) in multi-class EEG artifact classification. Proposed CNN models were kept as simple as possible to have the best operation time but in the meantime, models were selected adequately deep to extract appropriate artifact features from applied EEG signals. Obtained results prove that proposed architectures are able to classify artifacts with high accuracy.
{"title":"Deep Learning Approach for EEG Artifact Identification and Classification","authors":"R. Rajabioun, Ali Özen Akyürek, E. Sezer","doi":"10.1109/UBMK52708.2021.9558979","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558979","url":null,"abstract":"Electroencephalography (EEG) signals are normally susceptible to various artifacts and noises from different sources. In this paper, firstly the existence of artifacts will be identified on the recorded EEG signals and then the origin of the detected artifact will be determined among 7 different sources. Due to the nature of EEG signals, almost no specialist can determine artifact source through eye inspection. This paper introduces the utilization of 1-D Convolutional Neural Network (CNN) in multi-class EEG artifact classification. Proposed CNN models were kept as simple as possible to have the best operation time but in the meantime, models were selected adequately deep to extract appropriate artifact features from applied EEG signals. Obtained results prove that proposed architectures are able to classify artifacts with high accuracy.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123660371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558903
S. Taştan, G. Dalkılıç
Internet of things (IoT) has found its place in the modern world, and it continuously evolves. It is used in many areas varying from national defense systems to basic personal usages. One of the many areas that IoT devices are used is smart home systems. Smart home systems are gaining popularity because of their functionalities. They help users accomplish their tasks with swiftness and ease, which is a huge deal in busy lives of individuals. Pairing these already popular functionalities with more recent technologies such as Bluetooth, and Bluetooth Low Energy (BLE) mesh networks makes smart home systems even more useful. Using new versions of Bluetooth, which are specifically aimed at IoT applications, and a highly responsive topology like mesh significantly improves performance. In this project, the aim is to develop a mesh structure using Arduino integrated development environment (IDE). At the time, Arduino IDE does not support any BLE mesh libraries. Thus, to accomplish this, other libraries that provide connectivity to the chosen devices have been used. The functionalities needed to simulate a smart home system and the functionalities necessary for mesh networks have been implemented.
{"title":"Smart Home System Using Internet of Things Devices and Mesh Topology","authors":"S. Taştan, G. Dalkılıç","doi":"10.1109/UBMK52708.2021.9558903","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558903","url":null,"abstract":"Internet of things (IoT) has found its place in the modern world, and it continuously evolves. It is used in many areas varying from national defense systems to basic personal usages. One of the many areas that IoT devices are used is smart home systems. Smart home systems are gaining popularity because of their functionalities. They help users accomplish their tasks with swiftness and ease, which is a huge deal in busy lives of individuals. Pairing these already popular functionalities with more recent technologies such as Bluetooth, and Bluetooth Low Energy (BLE) mesh networks makes smart home systems even more useful. Using new versions of Bluetooth, which are specifically aimed at IoT applications, and a highly responsive topology like mesh significantly improves performance. In this project, the aim is to develop a mesh structure using Arduino integrated development environment (IDE). At the time, Arduino IDE does not support any BLE mesh libraries. Thus, to accomplish this, other libraries that provide connectivity to the chosen devices have been used. The functionalities needed to simulate a smart home system and the functionalities necessary for mesh networks have been implemented.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127678652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558986
Abdul Majeed Issifu, M. Ganiz
Easy Data Augmentation is originally developed for text classification tasks. It consists of four basic methods: Synonym Replacement, Random Insertion, Random Deletion, and Random Swap. They yield accuracy improvements on several deep neural network models. In this study we apply these methods to a new domain. We augment Named Entity Recognition datasets from medical domain. Although the augmentation task is much more difficult due to the nature of named entities which consist of word or word groups in the sentences, we show that we can improve the named entity recognition performance.
Easy Data Augmentation最初是为文本分类任务开发的。它包括四种基本方法:同义词替换、随机插入、随机删除和随机交换。它们在几个深度神经网络模型上提高了精度。在这项研究中，我们将这些方法应用到一个新的领域。我们增强了医学领域的命名实体识别数据集。尽管由于命名实体是由句子中的词或词组组成的，因此增强任务更加困难，但我们表明我们可以提高命名实体的识别性能。
{"title":"A Simple Data Augmentation Method to Improve the Performance of Named Entity Recognition Models in Medical Domain","authors":"Abdul Majeed Issifu, M. Ganiz","doi":"10.1109/UBMK52708.2021.9558986","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558986","url":null,"abstract":"Easy Data Augmentation is originally developed for text classification tasks. It consists of four basic methods: Synonym Replacement, Random Insertion, Random Deletion, and Random Swap. They yield accuracy improvements on several deep neural network models. In this study we apply these methods to a new domain. We augment Named Entity Recognition datasets from medical domain. Although the augmentation task is much more difficult due to the nature of named entities which consist of word or word groups in the sentences, we show that we can improve the named entity recognition performance.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129524038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：