{"title":"Parallel Toolkit for Measuring the Quality of Network Community Structure","authors":"Mingming Chen, Sisi Liu, B. Szymanski","doi":"10.1109/ENIC.2014.26","DOIUrl":"https://doi.org/10.1109/ENIC.2014.26","url":null,"abstract":"Many networks display community structure which identifies groups of nodes within which connections are denser than between them. Detecting and characterizing such community structure, which is known as community detection, is one of the fundamental issues in the study of network systems. It has received a considerable attention in the last years. Numerous techniques have been developed for both efficient and effective community detection. Among them, the most efficient algorithm is the label propagation algorithm whose computational complexity is O (|E|). Although it is linear in the number of edges, the running time is still too long for very large networks, creating the need for parallel community detection. Also, computing community quality metrics for community structure is computationally expensive both with and without ground truth. However, to date we are not aware of any effort to introduce parallelism for this problem. In this paper, we provide a parallel toolkit to calculate the values of such metrics. We evaluate the parallel algorithms on both distributed memory machine and shared memory machine. The experimental results show that they yield a significant performance gain over sequential execution in terms of total running time, speedup, and efficiency.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"604 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116369362","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}
{"title":"Topics and Sentiment Analysis of Evolving Social Groups","authors":"Bogdan Gliwa, Anna Zygmunt, J. Kozlak","doi":"10.1109/ENIC.2014.30","DOIUrl":"https://doi.org/10.1109/ENIC.2014.30","url":null,"abstract":"Groups are very significant elements of social networks, thus understanding their structure and reasons of their evolution are very important. In this paper a qualitative analysis of group evolution is proposed. The description of this process takes many aspects such as group transformation events, discussed topics and attitude of the discussion participants into consideration.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133271223","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}
{"title":"Durability of Links between Assets in Financial Markets: Minimal Spanning Trees and Correlations","authors":"A. Buda, A. Jarynowski","doi":"10.1109/ENIC.2014.18","DOIUrl":"https://doi.org/10.1109/ENIC.2014.18","url":null,"abstract":"We investigate hierarchical structure in stock markets according to Minimum Spanning Tree (MST) methods based on correlations between assets. The research was carried out on established (DJIA, DAX, FTSE100) and emerging markets (WIG 20). We consider durability of correlations between assets expressed by using the life-time of correlations between stocks or Minimum Spanning Trees half-life. In both methods durability of correlations depends on price history (time window width Δt). We extend our research on FOREX where the structure of Minimum Spanning Trees depends on basic currencies and reflects geographical connections. On the other hand, according to multistep survival ratio method, the survival of correlations and Minimum Spanning Trees does not depend on basic currency. We also detect a collective behavior and influences between single elements. The optimal window width used to compute correlation coefficients in financial markets is also discussed.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132673853","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}
{"title":"Searcher's Activity in Standalone and Web Applications as a Source for Search Query Expansion","authors":"Lubomir Vnenk, M. Bieliková","doi":"10.1109/ENIC.2014.19","DOIUrl":"https://doi.org/10.1109/ENIC.2014.19","url":null,"abstract":"The Web has so much variable information, therefore searching a specific one is complex task. To find something valuable, specifying good search query is crucial. However, average search query consists of only about two words which cannot specify intent of a searcher well. We suppose that these words follow in most cases the searcher's activity. Based on this assumption we propose an approach for search query extension by the searcher activity context. The activity context contains words reflecting the searcher's activity context expressed by keywords gathered from the content of one of the very recent activities within applications provided by the searcher. We present an approach for finding out searcher's activity context by analyzing the content and his interaction between applications considering both standalone applications and applications running inside a browser. Our method uses the activity logs to find a connection between the search query and specific application. Then we extend the query by terms gathered by analyzing the selected application content. We evaluate our approach in a series of experiments based on data gathered by monitoring a group of searchers by means of developed logger prototype.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117121399","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}
{"title":"Emotion Extraction from Turkish Text","authors":"Mansur Alp Toçoğlu, A. Alpkocak","doi":"10.1109/ENIC.2014.17","DOIUrl":"https://doi.org/10.1109/ENIC.2014.17","url":null,"abstract":"In this study we present an emotion extraction system from Turkish text. The system is able to recognizes even emotional states from a given text for happy, shame, guiltiness, disgust, sadness, angry and fear categories. We consider Emotion Extraction as a Text Classification problem, which requires a training set. Thus, we first obtained a survey which is done with 500 university students to develop a training set where they are asked to describe their most intense moments they remember for seven emotions categories. Then, the text describing emotional moments are pre processed and modeled in Vector Space Model where tf × idf weighting scheme is used. Then we applied Naive Bayes classifier and tested with 10-fold cross validation, in WEKA tool. We evaluated the system in terms of accuracy, precision, Measureand recall measures. The results we obtained from the first experimentation are very promising where it achieved around 86% accuracy for seven emotional classes in average.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130682117","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}
{"title":"Comparison of Scheduling Algorithms for Domain Specific Web Crawler","authors":"Krzysztof Filipowski","doi":"10.1109/ENIC.2014.14","DOIUrl":"https://doi.org/10.1109/ENIC.2014.14","url":null,"abstract":"Domain-specific Web crawlers are effective tools for acquiring information from the Web. One of the most crucial factors influencing the efficiency of domain crawlers is choice of crawling strategy. This article describes and compares several strategies for domain specific Web crawling. It concentrates particularly on scheduling algorithms which determine order of crawling URLs collected by the crawler. The objective of these strategies is to download the most relevant Web pages in an early stage of the crawl. In the paper there are presented four different algorithms which are compared using several metrics.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127845026","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}
K. Haniewicz, M. Kaczmarek, Magdalena Adamczyk, Wojciech Rutkowski
{"title":"A Case Study of Sentiment Orientation Identification for Polish Texts","authors":"K. Haniewicz, M. Kaczmarek, Magdalena Adamczyk, Wojciech Rutkowski","doi":"10.1109/ENIC.2014.25","DOIUrl":"https://doi.org/10.1109/ENIC.2014.25","url":null,"abstract":"In order to make rational decisions and react quickly to changes in the business environment, organizations, especially those operating in the e-business setting, need to constantly monitor numerous information sources on the Internet, e.g., electronic media or opinions published by users at various portals. Majority of the opinions and therefore data concerning enterprises is stored in various forms in broadly understood social media. The available resources are presented in a variety forms, both highly structured and free-style form. In order to identify the emotional attitude of the published texts, there is a well recognised need to automate the overall process and to employ sentiment analysis techniques. In this paper, we show how the already available resources for the Polish language, such as parsers, semantic networks, thesauri, general and domain specific corpora can be further extended and used as a cornerstone for more advanced applications. A case study of Sentiment Orientation Identification system for Polish texts is presented together with the obtained results from the conducted experiments.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124947010","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}
{"title":"MuSeNet: Collaboration in the Music Artists Industry","authors":"Alexandru Topîrceanu, Gabriel Barina, M. Udrescu","doi":"10.1109/ENIC.2014.10","DOIUrl":"https://doi.org/10.1109/ENIC.2014.10","url":null,"abstract":"Motivated by the constantly growing interest and real-world applicability shown in social networks, we model and analyze the network formed by music artists all around the world, which we call MuseNet. Inspired by similar approaches, we compare our analytic results with generic online friendship models and with the collaboration networks of actors. We are the first to fully create such a network, and by using centrality measures, we discover the most influential nodes in MuseNet. In light of current advances in social networks, we also highlight the importance of music producers in terms of meritocracy versus topological positioning, and discuss the differentiation between collaboration networks using a network fidelity approach. Finally, we show that MuseNet has a characteristic sociability -- a measure which is introduced in this paper - in comparison with other empirical networks.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132689250","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}
{"title":"Automated Classification of Scientific Collaborations with Network Indicators","authors":"Tilman Göhnert, A. Harrer, H. Hoppe","doi":"10.1109/ENIC.2014.20","DOIUrl":"https://doi.org/10.1109/ENIC.2014.20","url":null,"abstract":"In this paper we introduce a method and implementation to detect and classify specific episodes of scientific collaboration. Our method uses co-authorship networks and creates indicators for the discovery of temporal patterns of co-authoring. We apply the concept and implementation to scientific communities of the fields collaborative systems and social network analysis, to compare our findings to an earlier non-automated and small-scale analysis.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114786786","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}
{"title":"Sentiment Classification at the Time of the Tunisian Uprising: Machine Learning Techniques Applied to a New Corpus for Arabic Language","authors":"J. Akaichi","doi":"10.1109/ENIC.2014.35","DOIUrl":"https://doi.org/10.1109/ENIC.2014.35","url":null,"abstract":"Sentiment analysis is the field of study that analyzes people's opinions, sentiments, attitudes, and emotions from written language. It is one of the most active research areas in natural language processing and is also widely studied in data mining, web mining, and text mining. In recent years, text mining and sentiment analysis are being in almost every business and social domain which study all human activities and key influencers of our behaviors. Even though there are, at present, several studies related to this theme, most of them focus mainly on English texts. The resources available for opinion mining in other languages, such as Arabic, are still limited. In this paper, we propose a new sentiment analysis system destined to classify users' opinions which is performed with a new corpus for Arabic language gathered from users' posts at the time of the Tunisian revolution. Furthermore, different experiments have been carried out on this corpus, using machine learning algorithms such as Support Vector Machines and Naïve Bayes.","PeriodicalId":185148,"journal":{"name":"2014 European Network Intelligence Conference","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123874860","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}