際際滷shows by User: karpovilia / http://www.slideshare.net/images/logo.gif 際際滷shows by User: karpovilia / Thu, 15 Oct 2020 17:57:48 GMT 際際滷Share feed for 際際滷shows by User: karpovilia Detecting Automatically Managed Accounts in Online Social Networks: Graph Embedding Approach /slideshow/detecting-automatically-managed-accounts-in-online-social-networks-graph-embedding-approach/238888480 aist2020-201015175748
The widespread of Online Social Networks and the opportunity to commercialize popular accounts have attracted a large number of automated programs, known as artificial accounts. This paper focuses on the classification of human and fake accounts on the social network, by employing several graph neural networks, to efficiently encode attributes and network graph features of the account. Our work uses both network structure and attributes to distinguish human and artificial accounts and compares attributed and traditional graph embeddings. Separating complex, human-like artificial accounts into a standalone task demonstrates significant limitations of profile-based algorithms for bot detection and shows the efficiency of network structure-based methods for detecting sophisticated bot accounts. Experiments show that our approach can achieve competitive performance compared with existing state-of-the-art bot detection systems with only network-driven features. The source code of this paper is available at: http://github.com/karpovilia/botdetection.]]>
The widespread of Online Social Networks and the opportunity to commercialize popular accounts have attracted a large number of automated programs, known as artificial accounts. This paper focuses on the classification of human and fake accounts on the social network, by employing several graph neural networks, to efficiently encode attributes and network graph features of the account. Our work uses both network structure and attributes to distinguish human and artificial accounts and compares attributed and traditional graph embeddings. Separating complex, human-like artificial accounts into a standalone task demonstrates significant limitations of profile-based algorithms for bot detection and shows the efficiency of network structure-based methods for detecting sophisticated bot accounts. Experiments show that our approach can achieve competitive performance compared with existing state-of-the-art bot detection systems with only network-driven features. The source code of this paper is available at: http://github.com/karpovilia/botdetection.]]> Thu, 15 Oct 2020 17:57:48 GMT /slideshow/detecting-automatically-managed-accounts-in-online-social-networks-graph-embedding-approach/238888480 karpovilia@slideshare.net(karpovilia) Detecting Automatically Managed Accounts in Online Social Networks: Graph Embedding Approach karpovilia The widespread of Online Social Networks and the opportunity to commercialize popular accounts have attracted a large number of automated programs, known as artificial accounts. This paper focuses on the classification of human and fake accounts on the social network, by employing several graph neural networks, to efficiently encode attributes and network graph features of the account. Our work uses both network structure and attributes to distinguish human and artificial accounts and compares attributed and traditional graph embeddings. Separating complex, human-like artificial accounts into a standalone task demonstrates significant limitations of profile-based algorithms for bot detection and shows the efficiency of network structure-based methods for detecting sophisticated bot accounts. Experiments show that our approach can achieve competitive performance compared with existing state-of-the-art bot detection systems with only network-driven features. The source code of this paper is available at: http://github.com/karpovilia/botdetection. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/aist2020-201015175748-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> The widespread of Online Social Networks and the opportunity to commercialize popular accounts have attracted a large number of automated programs, known as artificial accounts. This paper focuses on the classification of human and fake accounts on the social network, by employing several graph neural networks, to efficiently encode attributes and network graph features of the account. Our work uses both network structure and attributes to distinguish human and artificial accounts and compares attributed and traditional graph embeddings. Separating complex, human-like artificial accounts into a standalone task demonstrates significant limitations of profile-based algorithms for bot detection and shows the efficiency of network structure-based methods for detecting sophisticated bot accounts. Experiments show that our approach can achieve competitive performance compared with existing state-of-the-art bot detection systems with only network-driven features. The source code of this paper is available at: http://github.com/karpovilia/botdetection.
Detecting Automatically Managed Accounts in Online Social Networks: Graph Embedding Approach from Ilia Karpov
]]> 350 0 https://cdn.slidesharecdn.com/ss_thumbnails/aist2020-201015175748-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 A general method applicable to the search for anglicisms in russian social network texts /slideshow/a-general-method-applicable-to-the-search-for-anglicisms-in-russian-social-network-texts/68926364 ageneralmethodapplicabletothesearchforanglicismsinrussiansocialnetworktexts-161114215233
In the process of globalization, the number of English words in other languages has rapidly increased. In automatic speech recognition systems, spell-checking, tagging, and other software in the field of natural language processing, loan words are not easily recognized and should be evaluated separately. In this paper we present a corpora-based approach to the automatic detection of anglicisms in Russian social network texts. Proposed method is based on the idea of simultaneous scripting, phonetics, and semantics similarity of the original Latin word and its Cyrillic analogue. We used a set of transliteration, phonetic transcribing, and morphological analysis methods to find possible hypotheses and distributional semantic models to filter them. Resulting list of borrowings, gathered from approximately 20 million LiveJournal texts, shows good intersection with manually collected dictionary. Proposed method is fully automated and can be applied to any domainspecific area. Full paper available at: https://www.academia.edu/29834070/A_General_Method_Applicable_to_the_Search_for_Anglicisms_in_Russian_Social_Network_Texts]]>
In the process of globalization, the number of English words in other languages has rapidly increased. In automatic speech recognition systems, spell-checking, tagging, and other software in the field of natural language processing, loan words are not easily recognized and should be evaluated separately. In this paper we present a corpora-based approach to the automatic detection of anglicisms in Russian social network texts. Proposed method is based on the idea of simultaneous scripting, phonetics, and semantics similarity of the original Latin word and its Cyrillic analogue. We used a set of transliteration, phonetic transcribing, and morphological analysis methods to find possible hypotheses and distributional semantic models to filter them. Resulting list of borrowings, gathered from approximately 20 million LiveJournal texts, shows good intersection with manually collected dictionary. Proposed method is fully automated and can be applied to any domainspecific area. Full paper available at: https://www.academia.edu/29834070/A_General_Method_Applicable_to_the_Search_for_Anglicisms_in_Russian_Social_Network_Texts]]> Mon, 14 Nov 2016 21:52:33 GMT /slideshow/a-general-method-applicable-to-the-search-for-anglicisms-in-russian-social-network-texts/68926364 karpovilia@slideshare.net(karpovilia) A general method applicable to the search for anglicisms in russian social network texts karpovilia In the process of globalization, the number of English words in other languages has rapidly increased. In automatic speech recognition systems, spell-checking, tagging, and other software in the field of natural language processing, loan words are not easily recognized and should be evaluated separately. In this paper we present a corpora-based approach to the automatic detection of anglicisms in Russian social network texts. Proposed method is based on the idea of simultaneous scripting, phonetics, and semantics similarity of the original Latin word and its Cyrillic analogue. We used a set of transliteration, phonetic transcribing, and morphological analysis methods to find possible hypotheses and distributional semantic models to filter them. Resulting list of borrowings, gathered from approximately 20 million LiveJournal texts, shows good intersection with manually collected dictionary. Proposed method is fully automated and can be applied to any domainspecific area. Full paper available at: https://www.academia.edu/29834070/A_General_Method_Applicable_to_the_Search_for_Anglicisms_in_Russian_Social_Network_Texts <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/ageneralmethodapplicabletothesearchforanglicismsinrussiansocialnetworktexts-161114215233-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> In the process of globalization, the number of English words in other languages has rapidly increased. In automatic speech recognition systems, spell-checking, tagging, and other software in the field of natural language processing, loan words are not easily recognized and should be evaluated separately. In this paper we present a corpora-based approach to the automatic detection of anglicisms in Russian social network texts. Proposed method is based on the idea of simultaneous scripting, phonetics, and semantics similarity of the original Latin word and its Cyrillic analogue. We used a set of transliteration, phonetic transcribing, and morphological analysis methods to find possible hypotheses and distributional semantic models to filter them. Resulting list of borrowings, gathered from approximately 20 million LiveJournal texts, shows good intersection with manually collected dictionary. Proposed method is fully automated and can be applied to any domainspecific area. Full paper available at: https://www.academia.edu/29834070/A_General_Method_Applicable_to_the_Search_for_Anglicisms_in_Russian_Social_Network_Texts
A general method applicable to the search for anglicisms in russian social network texts from Ilia Karpov
]]> 350 2 https://cdn.slidesharecdn.com/ss_thumbnails/ageneralmethodapplicabletothesearchforanglicismsinrussiansocialnetworktexts-161114215233-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 ENTITY BASED SENTIMENT ANALYSIS USING SYNTAX PATTERNS AND CONVOLUTIONAL NEURAL NETWORK鐃 /slideshow/entity-based-sentiment-analysis-using-syntax-patterns-and-convolutional-neural-network/62657991 kkkn-160602133651
This paper provides an alternative method to extracting object-based sentiment in text messages, based on modified method previously proposed by Mingbo [8], in which we first parse the syntax, and then correlate the sentiment with the object of analysis (also referred to as entity by some, therefore, used in this article interchangeably). We show two approaches for the sentiment polarity classification: syntactic rule patterns and convolutional neural network (CNN). Even without domain specific vocabulary and sophisticated classification algorithms, rule-based approach demonstrates an average macro-F1 based rank among the participants, whereas domain-specific vocabularies show a slightly higher macro-F1 score, but still close to an average result. CNN approach uses syntax dependencies and linear word order to obtain more extensive information about object relations. Convolution patterns, designed in this approach, are very similar to rules, obtained with rule-based approach. In our proposed approach, the neural network was trained with different Word2Vec (WV) models; we compared their performance relative to each other. In this paper, we show that learning a domain-specific WV offers slight progress in performance. Resulting macro-F1 score show performance in the into top three of the overall results among the competitors, participating in 2016 SentiRuEval event. Originally, we have not submitted our results to this competition at the time it was held, but had a chance to compare them post-hoc. We also combine the CNN approach with the rule-based approach and discuss the obtained differences in results. All training sets, evaluation metrics and experiments are used according to SentiRuEval 2016.]]>
This paper provides an alternative method to extracting object-based sentiment in text messages, based on modified method previously proposed by Mingbo [8], in which we first parse the syntax, and then correlate the sentiment with the object of analysis (also referred to as entity by some, therefore, used in this article interchangeably). We show two approaches for the sentiment polarity classification: syntactic rule patterns and convolutional neural network (CNN). Even without domain specific vocabulary and sophisticated classification algorithms, rule-based approach demonstrates an average macro-F1 based rank among the participants, whereas domain-specific vocabularies show a slightly higher macro-F1 score, but still close to an average result. CNN approach uses syntax dependencies and linear word order to obtain more extensive information about object relations. Convolution patterns, designed in this approach, are very similar to rules, obtained with rule-based approach. In our proposed approach, the neural network was trained with different Word2Vec (WV) models; we compared their performance relative to each other. In this paper, we show that learning a domain-specific WV offers slight progress in performance. Resulting macro-F1 score show performance in the into top three of the overall results among the competitors, participating in 2016 SentiRuEval event. Originally, we have not submitted our results to this competition at the time it was held, but had a chance to compare them post-hoc. We also combine the CNN approach with the rule-based approach and discuss the obtained differences in results. All training sets, evaluation metrics and experiments are used according to SentiRuEval 2016.]]> Thu, 02 Jun 2016 13:36:51 GMT /slideshow/entity-based-sentiment-analysis-using-syntax-patterns-and-convolutional-neural-network/62657991 karpovilia@slideshare.net(karpovilia) ENTITY BASED SENTIMENT ANALYSIS USING SYNTAX PATTERNS AND CONVOLUTIONAL NEURAL NETWORK鐃 karpovilia This paper provides an alternative method to extracting object-based sentiment in text messages, based on modified method previously proposed by Mingbo [8], in which we first parse the syntax, and then correlate the sentiment with the object of analysis (also referred to as entity by some, therefore, used in this article interchangeably). We show two approaches for the sentiment polarity classification: syntactic rule patterns and convolutional neural network (CNN). Even without domain specific vocabulary and sophisticated classification algorithms, rule-based approach demonstrates an average macro-F1 based rank among the participants, whereas domain-specific vocabularies show a slightly higher macro-F1 score, but still close to an average result. CNN approach uses syntax dependencies and linear word order to obtain more extensive information about object relations. Convolution patterns, designed in this approach, are very similar to rules, obtained with rule-based approach. In our proposed approach, the neural network was trained with different Word2Vec (WV) models; we compared their performance relative to each other. In this paper, we show that learning a domain-specific WV offers slight progress in performance. Resulting macro-F1 score show performance in the into top three of the overall results among the competitors, participating in 2016 SentiRuEval event. Originally, we have not submitted our results to this competition at the time it was held, but had a chance to compare them post-hoc. We also combine the CNN approach with the rule-based approach and discuss the obtained differences in results. All training sets, evaluation metrics and experiments are used according to SentiRuEval 2016. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/kkkn-160602133651-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> This paper provides an alternative method to extracting object-based sentiment in text messages, based on modified method previously proposed by Mingbo [8], in which we first parse the syntax, and then correlate the sentiment with the object of analysis (also referred to as entity by some, therefore, used in this article interchangeably). We show two approaches for the sentiment polarity classification: syntactic rule patterns and convolutional neural network (CNN). Even without domain specific vocabulary and sophisticated classification algorithms, rule-based approach demonstrates an average macro-F1 based rank among the participants, whereas domain-specific vocabularies show a slightly higher macro-F1 score, but still close to an average result. CNN approach uses syntax dependencies and linear word order to obtain more extensive information about object relations. Convolution patterns, designed in this approach, are very similar to rules, obtained with rule-based approach. In our proposed approach, the neural network was trained with different Word2Vec (WV) models; we compared their performance relative to each other. In this paper, we show that learning a domain-specific WV offers slight progress in performance. Resulting macro-F1 score show performance in the into top three of the overall results among the competitors, participating in 2016 SentiRuEval event. Originally, we have not submitted our results to this competition at the time it was held, but had a chance to compare them post-hoc. We also combine the CNN approach with the rule-based approach and discuss the obtained differences in results. All training sets, evaluation metrics and experiments are used according to SentiRuEval 2016.
ENTITY BASED SENTIMENT ANALYSIS USING SYNTAX PATTERNS AND CONVOLUTIONAL NEURAL NETWORK from Ilia Karpov
]]> 263 5 https://cdn.slidesharecdn.com/ss_thumbnails/kkkn-160602133651-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 亳从舒 亳 亳仆亠亞舒亳 仂从 亟舒仆仆 亳亰 舒亰仆仂仂亟仆 亳仂仆亳从仂于 /slideshow/ss-56145040/56145040 random-151215002612
仂弍亰仂 仆亠从仂仂 仗仂亟仂亟仂于 从 亠亠仆亳 仗仂弍仍亠仄 data curation. 仂从仂仍从 亟仂仍仂于仆 从舒仍从 亟亠仍舒 仆亠 仂亠仍仂, 亳仗仂仍亰仂于舒仍 亠仄亳仆 "仂亳从舒 亳 亳仆亠亞舒亳".]]>
仂弍亰仂 仆亠从仂仂 仗仂亟仂亟仂于 从 亠亠仆亳 仗仂弍仍亠仄 data curation. 仂从仂仍从 亟仂仍仂于仆 从舒仍从 亟亠仍舒 仆亠 仂亠仍仂, 亳仗仂仍亰仂于舒仍 亠仄亳仆 "仂亳从舒 亳 亳仆亠亞舒亳".]]> Tue, 15 Dec 2015 00:26:12 GMT /slideshow/ss-56145040/56145040 karpovilia@slideshare.net(karpovilia) 亳从舒 亳 亳仆亠亞舒亳 仂从 亟舒仆仆 亳亰 舒亰仆仂仂亟仆 亳仂仆亳从仂于 karpovilia 仂弍亰仂 仆亠从仂仂 仗仂亟仂亟仂于 从 亠亠仆亳 仗仂弍仍亠仄 data curation. 仂从仂仍从 亟仂仍仂于仆 从舒仍从 亟亠仍舒 仆亠 仂亠仍仂, 亳仗仂仍亰仂于舒仍 亠仄亳仆 "仂亳从舒 亳 亳仆亠亞舒亳". <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/random-151215002612-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> 仂弍亰仂 仆亠从仂仂 仗仂亟仂亟仂于 从 亠亠仆亳 仗仂弍仍亠仄 data curation. 仂从仂仍从 亟仂仍仂于仆 从舒仍从 亟亠仍舒 仆亠 仂亠仍仂, 亳仗仂仍亰仂于舒仍 亠仄亳仆 &quot;仂亳从舒 亳 亳仆亠亞舒亳&quot;.
亳从舒 亳 亳仆亠亞舒亳 仂从 亟舒仆仆 亳亰 舒亰仆仂仂亟仆 亳仂仆亳从仂于 from Ilia Karpov
]]> 303 4 https://cdn.slidesharecdn.com/ss_thumbnails/random-151215002612-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 弌弍仂, 舒仆舒仍亳亰, 仂弍舒弍仂从舒 亠从仂于仂亶 亳仆仂仄舒亳亳 /slideshow/ss-28646925/28646925 random-131126125713-phpapp01
亠亰亠仆舒亳 亟仂从仍舒亟舒 仆舒 亠仄亳仆舒亠 "弌亠亠于亠 仄亠仂亟 亳 仄仂亟亠仍亳 于 舒仆舒仍亳亰亠 亠从仂于仂亶 亳仆仂仄舒亳亳" 于 丿亅]]>
亠亰亠仆舒亳 亟仂从仍舒亟舒 仆舒 亠仄亳仆舒亠 "弌亠亠于亠 仄亠仂亟 亳 仄仂亟亠仍亳 于 舒仆舒仍亳亰亠 亠从仂于仂亶 亳仆仂仄舒亳亳" 于 丿亅]]> Tue, 26 Nov 2013 12:57:13 GMT /slideshow/ss-28646925/28646925 karpovilia@slideshare.net(karpovilia) 弌弍仂, 舒仆舒仍亳亰, 仂弍舒弍仂从舒 亠从仂于仂亶 亳仆仂仄舒亳亳 karpovilia 亠亰亠仆舒亳 亟仂从仍舒亟舒 仆舒 亠仄亳仆舒亠 "弌亠亠于亠 仄亠仂亟 亳 仄仂亟亠仍亳 于 舒仆舒仍亳亰亠 亠从仂于仂亶 亳仆仂仄舒亳亳" 于 丿亅 <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/random-131126125713-phpapp01-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> 亠亰亠仆舒亳 亟仂从仍舒亟舒 仆舒 亠仄亳仆舒亠 &quot;弌亠亠于亠 仄亠仂亟 亳 仄仂亟亠仍亳 于 舒仆舒仍亳亰亠 亠从仂于仂亶 亳仆仂仄舒亳亳&quot; 于 丿亅
弌弍仂, 舒仆舒仍亳亰, 仂弍舒弍仂从舒 亠从仂于仂亶 亳仆仂仄舒亳亳 from Ilia Karpov
]]> 1402 2 https://cdn.slidesharecdn.com/ss_thumbnails/random-131126125713-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 https://public.slidesharecdn.com/v2/images/profile-picture.png https://cdn.slidesharecdn.com/ss_thumbnails/aist2020-201015175748-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/detecting-automatically-managed-accounts-in-online-social-networks-graph-embedding-approach/238888480 Detecting Automaticall... https://cdn.slidesharecdn.com/ss_thumbnails/ageneralmethodapplicabletothesearchforanglicismsinrussiansocialnetworktexts-161114215233-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/a-general-method-applicable-to-the-search-for-anglicisms-in-russian-social-network-texts/68926364 A general method appli... https://cdn.slidesharecdn.com/ss_thumbnails/kkkn-160602133651-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/entity-based-sentiment-analysis-using-syntax-patterns-and-convolutional-neural-network/62657991 ENTITY BASED SENTIMENT...