Pub Date : 2018-11-28DOI: 10.21533/SCJOURNAL.V7I2.158
M. Can, Osman Gursoy
In 1980s, the face of the microbiology dramatically changed with the rRNA-based phylogenetic classifications, by Carl Woese. He delineated the three main branches of life. He used the technique not only to explore microbial diversity but also as a method for bacterial annotation. Today, rRNA-based analysis remains a central method in microbiology. Many researchers followed this track, using several new generations of Artificial Neural Networks they obtained high accuracies using available datasets of their time. Recently the number of known bacteria increased enormously. In this article we used ANN's to annotate bacterial 16S rRNA gene sequences from five selected phylums in Greengenes database taxonomy: Proteobacteria, Firmicutes, Bacteroidetes, Actinobacteria, and Chloroflexi. 93% average accuracy is obtained in classif-ications. When we used the bundle testing technique, the average accuracy easily raised to 100%.
{"title":"Artificial Neural Networks in Bacteria Taxonomic Classification","authors":"M. Can, Osman Gursoy","doi":"10.21533/SCJOURNAL.V7I2.158","DOIUrl":"https://doi.org/10.21533/SCJOURNAL.V7I2.158","url":null,"abstract":"In 1980s, the face of the microbiology dramatically changed with the rRNA-based phylogenetic classifications, by Carl Woese. He delineated the three main branches of life. He used the technique not only to explore microbial diversity but also as a method for bacterial annotation. Today, rRNA-based analysis remains a central method in microbiology. Many researchers followed this track, using several new generations of Artificial Neural Networks they obtained high accuracies using available datasets of their time. Recently the number of known bacteria increased enormously. In this article we used ANN's to annotate bacterial 16S rRNA gene sequences from five selected phylums in Greengenes database taxonomy: Proteobacteria, Firmicutes, Bacteroidetes, Actinobacteria, and Chloroflexi. 93% average accuracy is obtained in classif-ications. When we used the bundle testing technique, the average accuracy easily raised to 100%.","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114428398","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 : 2018-11-28DOI: 10.21533/scjournal.v7i2.164
J. Karabegović
Changes and fluctuations in commodity prices exert different effects on value chain participants, depending on the position they have in the chain. Agricultural commodities are exposed to a set of different factors influencing the prices of the commodities. They are influenced by the season, weather shocks, demand and supply forces, household income, tastes and preferences of the consumers. Observing the most recent history, high price fluctuations have been observed during the financial crisis in 2008. One out of many approaches for hedging the price risk is the usage of financial derivatives. This study will be concerned with the volatility modelling methods with the help of futures and options for corn and soya. Methods used for modelling the volatility a Black Scholes Implied Volatility. The simplest method in ARCH family, namely the GARCH (1,1) method will be used for modelling volatility based on the historical futures data dating back to 2005. The implied volatility is derived solving back the Black – Scholes Model, only this time looking for sigma. The sole purpose of the thesis is to examine which of the two methods has a better predictive power. Model comparison is done with the help of forecast regression models. The regression models have shown the difficulty in assessing which model has more accurate predictive power. The Adjusted R2 for both models in both cases is relatively low. However, the GARCH (1,1) model has slightly higher values for this indicator. Even the GARCH (1,1) model h a better performance, due to the relatively low adjusted R2 values, no stable conclusion regarding the model performance can be derived. assets(Hull, 2008). Options are financial instruments in the form of a contract that are traded between the writer of an option and an option holder, and it provides the re GARCH and
{"title":"Risk Exposition of Prices in Agricultural Commodities Using Options and Futures","authors":"J. Karabegović","doi":"10.21533/scjournal.v7i2.164","DOIUrl":"https://doi.org/10.21533/scjournal.v7i2.164","url":null,"abstract":"Changes and fluctuations in commodity prices exert different effects on value chain participants, depending on the position they have in the chain. Agricultural commodities are exposed to a set of different factors influencing the prices of the commodities. They are influenced by the season, weather shocks, demand and supply forces, household income, tastes and preferences of the consumers. Observing the most recent history, high price fluctuations have been observed during the financial crisis in 2008. One out of many approaches for hedging the price risk is the usage of financial derivatives. This study will be concerned with the volatility modelling methods with the help of futures and options for corn and soya. Methods used for modelling the volatility a Black Scholes Implied Volatility. The simplest method in ARCH family, namely the GARCH (1,1) method will be used for modelling volatility based on the historical futures data dating back to 2005. The implied volatility is derived solving back the Black – Scholes Model, only this time looking for sigma. The sole purpose of the thesis is to examine which of the two methods has a better predictive power. Model comparison is done with the help of forecast regression models. The regression models have shown the difficulty in assessing which model has more accurate predictive power. The Adjusted R2 for both models in both cases is relatively low. However, the GARCH (1,1) model has slightly higher values for this indicator. Even the GARCH (1,1) model h a better performance, due to the relatively low adjusted R2 values, no stable conclusion regarding the model performance can be derived. assets(Hull, 2008). Options are financial instruments in the form of a contract that are traded between the writer of an option and an option holder, and it provides the re GARCH and","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"152 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122429473","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 : 2018-11-28DOI: 10.21533/SCJOURNAL.V7I2.160
A. Abidi, M. Can
{"title":"Validation Tools for Predicted Linear B-Epitopes: Antigenicity","authors":"A. Abidi, M. Can","doi":"10.21533/SCJOURNAL.V7I2.160","DOIUrl":"https://doi.org/10.21533/SCJOURNAL.V7I2.160","url":null,"abstract":"","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116962272","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 : 2018-11-28DOI: 10.21533/SCJOURNAL.V7I2.163
N. M. Demir
Dataset consists of 17000 tweets collected from Twitter, as 500 tweets for each of 34 authors that meet certain criteria. Raw data is collected by using the software Nvivo. The collected raw data is preprocessed to extract frequencies of 200 features. In the data analysis 128 of features are eliminated since they are rare in tweets. As a progressive presentation, five – ten – fifteen – twenty - thirty and thirty four of these 34 authors are selected each time. Since recurrent artificial neural networks are more stable and iterations converge more quickly, in this work this architecture is preferred. In general, ANNs are more successful in distinguishing two classes, therefore for N authors, N×N neural networks are trained for pair wise classification. These N×N experts then organized as N special teams (CANNT) to aggregate decisions of these N×N experts. Number of authors is seen not so effective on the accuracy of the authentication, and around 80% accuracy is achieved for any number of authors.
{"title":"Authorship Authentication of Short Messages from Social Networks Using Recurrent Artificial Neural Networks","authors":"N. M. Demir","doi":"10.21533/SCJOURNAL.V7I2.163","DOIUrl":"https://doi.org/10.21533/SCJOURNAL.V7I2.163","url":null,"abstract":"Dataset consists of 17000 tweets collected from Twitter, as 500 tweets for each of 34 authors that meet certain criteria. Raw data is collected by using the software Nvivo. The collected raw data is preprocessed to extract frequencies of 200 features. In the data analysis 128 of features are eliminated since they are rare in tweets. As a progressive presentation, five – ten – fifteen – twenty - thirty and thirty four of these 34 authors are selected each time. Since recurrent artificial neural networks are more stable and iterations converge more quickly, in this work this architecture is preferred. In general, ANNs are more successful in distinguishing two classes, therefore for N authors, N×N neural networks are trained for pair wise classification. These N×N experts then organized as N special teams (CANNT) to aggregate decisions of these N×N experts. Number of authors is seen not so effective on the accuracy of the authentication, and around 80% accuracy is achieved for any number of authors.","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134340486","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 : 2018-11-28DOI: 10.21533/SCJOURNAL.V7I2.166
M. Can, Osman Gursoy
In 1980s, Carl Woese made a ground breaking contribution to microbiology using rRNA-genes for phylogenetic classifications. He used it not only to explore microbial diversity but also as a method for bacterial annotation. Today, rRNA-based analysis remains a central method in microbiology. Many researchers followed this track, using several new generations of Artificial Neural Networks obtained high accuracies using available datasets of their time. By the time, the number of bacteria increased enormously. In this article we used Longest Common Subsequence similarity measure to classify bacterial 16S rRNA gene sequences of 1.820.414 bacteria in SILVA, 3.196.038 bacteria in RDP, and 198.509 bacteria in Greengenes. The last two taxonomy have six taxonomical levels, phylum, class, order, family, genus, and species, while SILVA has two more levels subclass and suborder, but lacks species level. The majority of classifications (98%) were of high accuracy (98%).
{"title":"Longest Common Subsequences in Bacteria Taxonomic Classification","authors":"M. Can, Osman Gursoy","doi":"10.21533/SCJOURNAL.V7I2.166","DOIUrl":"https://doi.org/10.21533/SCJOURNAL.V7I2.166","url":null,"abstract":"In 1980s, Carl Woese made a ground breaking contribution to microbiology using rRNA-genes for phylogenetic classifications. He used it not only to explore microbial diversity but also as a method for bacterial annotation. Today, rRNA-based analysis remains a central method in microbiology. Many researchers followed this track, using several new generations of Artificial Neural Networks obtained high accuracies using available datasets of their time. By the time, the number of bacteria increased enormously. In this article we used Longest Common Subsequence similarity measure to classify bacterial 16S rRNA gene sequences of 1.820.414 bacteria in SILVA, 3.196.038 bacteria in RDP, and 198.509 bacteria in Greengenes. The last two taxonomy have six taxonomical levels, phylum, class, order, family, genus, and species, while SILVA has two more levels subclass and suborder, but lacks species level. The majority of classifications (98%) were of high accuracy (98%).","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124142091","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 : 2018-11-28DOI: 10.21533/SCJOURNAL.V7I2.162
A. Abidi
{"title":"Validation Tools for Predicted Linear B-Epitopes: Beta Turns","authors":"A. Abidi","doi":"10.21533/SCJOURNAL.V7I2.162","DOIUrl":"https://doi.org/10.21533/SCJOURNAL.V7I2.162","url":null,"abstract":"","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114562020","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 : 2018-11-28DOI: 10.21533/SCJOURNAL.V7I2.161
R. Gosto
Protein 3-dimensional structure prediction is determination of the 3-dimensional structure of a protein from its amino acid sequence by using protein structure prediction software. By understanding protein’s 3-dimensional structure, we should be able to figure out the function of the said protein. We already have several protein prediction software, but the purpose of this study is to determine how accurate they are, and if the results presented are true and to what extent. To determine how accurate protein 3-dimensional structure prediction software are, we compared x-ray crystallography determined protein structures to software predicted 3-dimensonal protein structures. All of the software used showed good accuracy, and according to our results, “i-Tasser” software was the most accurate, closely followed by RaptorX.
{"title":"Assessment of Accuracies of Protein 3-Dimensional Prediction Software","authors":"R. Gosto","doi":"10.21533/SCJOURNAL.V7I2.161","DOIUrl":"https://doi.org/10.21533/SCJOURNAL.V7I2.161","url":null,"abstract":"Protein 3-dimensional structure prediction is determination of the 3-dimensional structure of a protein from its amino acid sequence by using protein structure prediction software. By understanding protein’s 3-dimensional structure, we should be able to figure out the function of the said protein. We already have several protein prediction software, but the purpose of this study is to determine how accurate they are, and if the results presented are true and to what extent. To determine how accurate protein 3-dimensional structure prediction software are, we compared x-ray crystallography determined protein structures to software predicted 3-dimensonal protein structures. All of the software used showed good accuracy, and according to our results, “i-Tasser” software was the most accurate, closely followed by RaptorX.","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132789796","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 : 2018-11-28DOI: 10.21533/SCJOURNAL.V7I2.159
A. Selimi, M. Saračević
Computational geometry is an integral part of mathematics and computer science deals with the algorithmic solution of geometry problems. From the beginning to today, computer geometry links different areas of science and techniques, such as the theory of algorithms, combinatorial and Euclidean geometry, but including data structures and optimization. Today, computational geometry has a great deal of application in computer graphics, geometric modeling, computer vision, and geodesic path, motion planning and parallel computing. The complex calculations and theories in the field of geometry are long time studied and developed, but from the aspect of application in modern information technologies they still are in the beginning. In this research is given the applications of computational geometry in polygon triangulation, manufacturing of objects with molds, point location, and robot motion planning.
{"title":"Computational Geometry Applications","authors":"A. Selimi, M. Saračević","doi":"10.21533/SCJOURNAL.V7I2.159","DOIUrl":"https://doi.org/10.21533/SCJOURNAL.V7I2.159","url":null,"abstract":"Computational geometry is an integral part of mathematics and computer science deals with the algorithmic solution of geometry problems. From the beginning to today, computer geometry links different areas of science and techniques, such as the theory of algorithms, combinatorial and Euclidean geometry, but including data structures and optimization. Today, computational geometry has a great deal of application in computer graphics, geometric modeling, computer vision, and geodesic path, motion planning and parallel computing. The complex calculations and theories in the field of geometry are long time studied and developed, but from the aspect of application in modern information technologies they still are in the beginning. In this research is given the applications of computational geometry in polygon triangulation, manufacturing of objects with molds, point location, and robot motion planning.","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133864138","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 : 2018-11-28DOI: 10.21533/SCJOURNAL.V7I2.165
A. Abidi
Experimental methods used for characterizing epitopes that play a vital role in the development of peptide vaccines, in diagnosis of diseases, and also for allergy research are time consuming and need huge resources. There are many online epitope prediction tools are available that can help scientists in short listing the candidate peptides. To predict B-cell epitopes in an antigenic sequence, Jordan recurrent neural network (BIRUNI) is found to besuccessful. To train and test neural networks, 262.583 B epitopes are retrieved from IEDB database. 99.9% of these epitopes have lengths in the interval 6-25 amino acids. For each of these lengths, committees of 11 expert recurrent neural networks are trained. To train these experts alongside epitopes, non-epitopes are needed. Non-epitopes are created as random sequences of amino acids of the same length followed by a filtering process. To distinguish epitopes and non-epitopes, the votes of eleven experts are aggregated by majority vote. An overall accuracy of 97.23% is achieved. Then these experts are used to predict the Linear Bepitopes of five antigens, Plasmodium Falciparum, Human Polio Virus Sabin Strain, Meningitis, Plasmodium Vivax and Mycobacterium Tuberculosis. The success of BIRUNU is compared with the five online prediction tools ABCPRED, BCPRED, K&T, BEPIPRED, and AAP.It is seen that BIRUNI outperforms all of them in the average.
{"title":"A Recurrent Neural Network Linear B-Epitope Predictor: BIRUNI","authors":"A. Abidi","doi":"10.21533/SCJOURNAL.V7I2.165","DOIUrl":"https://doi.org/10.21533/SCJOURNAL.V7I2.165","url":null,"abstract":"Experimental methods used for characterizing epitopes that play a vital role in the development of peptide vaccines, in diagnosis of diseases, and also for allergy research are time consuming and need huge resources. There are many online epitope prediction tools are available that can help scientists in short listing the candidate peptides. To predict B-cell epitopes in an antigenic sequence, Jordan recurrent neural network (BIRUNI) is found to besuccessful. To train and test neural networks, 262.583 B epitopes are retrieved from IEDB database. 99.9% of these epitopes have lengths in the interval 6-25 amino acids. For each of these lengths, committees of 11 expert recurrent neural networks are trained. To train these experts alongside epitopes, non-epitopes are needed. Non-epitopes are created as random sequences of amino acids of the same length followed by a filtering process. To distinguish epitopes and non-epitopes, the votes of eleven experts are aggregated by majority vote. An overall accuracy of 97.23% is achieved. Then these experts are used to predict the Linear Bepitopes of five antigens, Plasmodium Falciparum, Human Polio Virus Sabin Strain, Meningitis, Plasmodium Vivax and Mycobacterium Tuberculosis. The success of BIRUNU is compared with the five online prediction tools ABCPRED, BCPRED, K&T, BEPIPRED, and AAP.It is seen that BIRUNI outperforms all of them in the average.","PeriodicalId":243185,"journal":{"name":"Southeast Europe Journal of Soft Computing","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126882441","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：