Pub Date : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974483
Amandeep Kaur, D. Chahal, LATIKA KHARB
With the advent of LPG Policy many foreign firms have invested in India and home country firms too have expanded their operations outside the nation which results in greater inflow and outflow of foreign currency and INR giving rise to currency fluctuations, these fluctuations were exposed to more volatility after the US financial crisis of 2008. The increased risk exposure can be mitigated by employing techniques such as hedging, speculation and by using instruments like Futures and Forward contracts. Currency futures have been considered as the best instruments for managing risk against foreign currency exchange rate volatility. Keeping this into consideration the present paper analyses the efficiency of random walk hypothesis by testing currency futures in weak form efficiency post financial crisis of 2008. The sample includes the monthly closing price indices for the period January 2009 to March 2019. The hypothesis tested is whether the currency futures USD/INR are weak form efficient. Statistical tools employed in the study encompasses Runs Test, Autocorrelation Function, Kolmogorov-Smirnov Test (K-S Test), Augmented Dickey Fuller Test (ADF Test), and Ljung Box Test. The results of the employed tests provide evidence on the non-randomness of the time series.
{"title":"Weak Form Efficiency Of Currency Futures: Evidence From India","authors":"Amandeep Kaur, D. Chahal, LATIKA KHARB","doi":"10.1109/ICCCIS48478.2019.8974483","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974483","url":null,"abstract":"With the advent of LPG Policy many foreign firms have invested in India and home country firms too have expanded their operations outside the nation which results in greater inflow and outflow of foreign currency and INR giving rise to currency fluctuations, these fluctuations were exposed to more volatility after the US financial crisis of 2008. The increased risk exposure can be mitigated by employing techniques such as hedging, speculation and by using instruments like Futures and Forward contracts. Currency futures have been considered as the best instruments for managing risk against foreign currency exchange rate volatility. Keeping this into consideration the present paper analyses the efficiency of random walk hypothesis by testing currency futures in weak form efficiency post financial crisis of 2008. The sample includes the monthly closing price indices for the period January 2009 to March 2019. The hypothesis tested is whether the currency futures USD/INR are weak form efficient. Statistical tools employed in the study encompasses Runs Test, Autocorrelation Function, Kolmogorov-Smirnov Test (K-S Test), Augmented Dickey Fuller Test (ADF Test), and Ljung Box Test. The results of the employed tests provide evidence on the non-randomness of the time series.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"21 23","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114049965","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974528
Chinkit Manchanda, Rajat Rathi, Nikhil Sharma
Traffic congestion is a common affair in the big cities and towns. This issue is the outcome of the rapid increase in the population and increasing number of vehicles, so predicting the level of traffic congestion will be beneficial for every individual. However, interpretation and implementation of traffic state can be exceptionally tough. With this pace of increasing vehicles, existing algorithms may come up with some limitations due to various aspects of features which we cannot process. In this paper, we introduce a Hybrid Deep Neural Network (HDNN) for forecasting the traffic conditions on roads with the images using Convolutional Neural Network (CNN) and predicting road accident statistics of a particular area on a specific time. This model will exploit the development of algorithms in machine learning and majorly grasping over the Deep learning algorithm CNN. Experimental results show superior results of traffic conditions prediction and road accidentsanalysis, HDNN outshine the standard benchmark for the level of traffic congestion.
{"title":"Traffic Density Investigation & Road Accident Analysis in India using Deep Learning","authors":"Chinkit Manchanda, Rajat Rathi, Nikhil Sharma","doi":"10.1109/ICCCIS48478.2019.8974528","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974528","url":null,"abstract":"Traffic congestion is a common affair in the big cities and towns. This issue is the outcome of the rapid increase in the population and increasing number of vehicles, so predicting the level of traffic congestion will be beneficial for every individual. However, interpretation and implementation of traffic state can be exceptionally tough. With this pace of increasing vehicles, existing algorithms may come up with some limitations due to various aspects of features which we cannot process. In this paper, we introduce a Hybrid Deep Neural Network (HDNN) for forecasting the traffic conditions on roads with the images using Convolutional Neural Network (CNN) and predicting road accident statistics of a particular area on a specific time. This model will exploit the development of algorithms in machine learning and majorly grasping over the Deep learning algorithm CNN. Experimental results show superior results of traffic conditions prediction and road accidentsanalysis, HDNN outshine the standard benchmark for the level of traffic congestion.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127502250","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974537
V. Rohilla, Ms Sanika Singh kumar, Sudeshna Chakraborty, Ms. Sanika Singh
Clustering is one the one of the most important technique of data mining. It is used in many applications like fraud detection, image processing, bioinformatics etc. It has been used in various domains. Many types of the clustering techniques are the following like hierarchical, partitional, spectral clustering, density clustering, grid clustering, model based clustering etc. Bisecting K-Means comes under partitional clustering. It gives better performane, when huge data is used. There are many approached that are developed in the similar domain.One of the technique is Text Mining through which useful information is extracted through text. One of the important concept is statistical pattern mining through which important information is extracted by planning different trends and patterns. Input text patterns are structured that are derived from structured data and corresponding output is generated. The steps of text mining are categories of text, clustering text, extraction, summarization of text, E-R modeling. The various steps of text analysis are retrieval of information, lex. analysis for distribution of word freq. distribution study, recognition of pattern,tagging, extraction of information, techniques of data mining and also link analysis, association, visual. and predictive analyt. In the given paper bisect. K Means algorithm is presented which has the features of k-Means and hierar. clustering.
{"title":"Data Clustering using Bisecting K-Means","authors":"V. Rohilla, Ms Sanika Singh kumar, Sudeshna Chakraborty, Ms. Sanika Singh","doi":"10.1109/ICCCIS48478.2019.8974537","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974537","url":null,"abstract":"Clustering is one the one of the most important technique of data mining. It is used in many applications like fraud detection, image processing, bioinformatics etc. It has been used in various domains. Many types of the clustering techniques are the following like hierarchical, partitional, spectral clustering, density clustering, grid clustering, model based clustering etc. Bisecting K-Means comes under partitional clustering. It gives better performane, when huge data is used. There are many approached that are developed in the similar domain.One of the technique is Text Mining through which useful information is extracted through text. One of the important concept is statistical pattern mining through which important information is extracted by planning different trends and patterns. Input text patterns are structured that are derived from structured data and corresponding output is generated. The steps of text mining are categories of text, clustering text, extraction, summarization of text, E-R modeling. The various steps of text analysis are retrieval of information, lex. analysis for distribution of word freq. distribution study, recognition of pattern,tagging, extraction of information, techniques of data mining and also link analysis, association, visual. and predictive analyt. In the given paper bisect. K Means algorithm is presented which has the features of k-Means and hierar. clustering.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128495219","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974505
S. Parihar, S. Borana, S. K. Yadav
Remote sensing data increasingly used to identify and map open water bodies at comprehensive scales. The Jodhpur city experienced high urbanization in last one decade with population about 13Lakhs, and therefore needs effective public amenities alongwith safe water. In this research, compared accessible surface water bodies mapping approaches using seven spectral indices, viz. normalized difference water index (NDWI), tasseled cap wetness index (TCW), automated water extraction index (AWEIsh and AWEInsh), modified normalized difference water index (MNDWI), Water ratio index (WRI) and normalized difference vegetation index (NDVI) as well as two medium resolution sensors (Sentinel-2A and Landsat 8 OLI). The combinations of different water algorithms and satellite sensors were used to evaluate accuracy of the open water body. The results confirmed that water algorithms have high accuracies with Kappa Coefficients ranging from 0.12 to 0.98. The MNDWI water algorithms performed better than other water indices algorithms, and could be associated with pure water dominance in study area.The resultant water mapping from Sentinel-2A (10m) data has superior accuracies than Sentinel-2 (20m) and Landsat 8 OLI (30m). This research illustrates the enhanced performance in Sentinel-2A (10m) and Sentinel-2A (20m) for mapping of water body. The present study shows the availability of alternate water resources, which shall also be useful during frequent maintenance work of Rajeev Gandhi canal which supply water to whole city. The nearest neighbour technique is used to resample Sentibel-2A (10m) data of the visible and near IR bands to 20 m resolution bands to perform further analysis and comparisons.
{"title":"Comparative Evaluation of Spectral Indices and Sensors for Mapping of Urban Surface Water Bodies in Jodhpur Area : Smart & Sustainable Growth","authors":"S. Parihar, S. Borana, S. K. Yadav","doi":"10.1109/ICCCIS48478.2019.8974505","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974505","url":null,"abstract":"Remote sensing data increasingly used to identify and map open water bodies at comprehensive scales. The Jodhpur city experienced high urbanization in last one decade with population about 13Lakhs, and therefore needs effective public amenities alongwith safe water. In this research, compared accessible surface water bodies mapping approaches using seven spectral indices, viz. normalized difference water index (NDWI), tasseled cap wetness index (TCW), automated water extraction index (AWEIsh and AWEInsh), modified normalized difference water index (MNDWI), Water ratio index (WRI) and normalized difference vegetation index (NDVI) as well as two medium resolution sensors (Sentinel-2A and Landsat 8 OLI). The combinations of different water algorithms and satellite sensors were used to evaluate accuracy of the open water body. The results confirmed that water algorithms have high accuracies with Kappa Coefficients ranging from 0.12 to 0.98. The MNDWI water algorithms performed better than other water indices algorithms, and could be associated with pure water dominance in study area.The resultant water mapping from Sentinel-2A (10m) data has superior accuracies than Sentinel-2 (20m) and Landsat 8 OLI (30m). This research illustrates the enhanced performance in Sentinel-2A (10m) and Sentinel-2A (20m) for mapping of water body. The present study shows the availability of alternate water resources, which shall also be useful during frequent maintenance work of Rajeev Gandhi canal which supply water to whole city. The nearest neighbour technique is used to resample Sentibel-2A (10m) data of the visible and near IR bands to 20 m resolution bands to perform further analysis and comparisons.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122127749","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974553
Amit Kumar Yadav, R. Johari, Raman Dahiya
World have many complex systems and each of them composed of various smaller components. Network science has been used to study the complex system like Brain with numerous neurons, Internet, Business Connection and within other domains. This paper focused on study about Network Science, Social Network Analysis and identifying centrality measures using literature survey. In further process R code had applied to find Centrality measures of sample network and Community detection is done on real dataset gathered from USDS (USICT Student Dataset) using Network Science, Newman-Girvan Algorithm.
{"title":"Identification of Centrality Measures in Social Network using Network Science","authors":"Amit Kumar Yadav, R. Johari, Raman Dahiya","doi":"10.1109/ICCCIS48478.2019.8974553","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974553","url":null,"abstract":"World have many complex systems and each of them composed of various smaller components. Network science has been used to study the complex system like Brain with numerous neurons, Internet, Business Connection and within other domains. This paper focused on study about Network Science, Social Network Analysis and identifying centrality measures using literature survey. In further process R code had applied to find Centrality measures of sample network and Community detection is done on real dataset gathered from USDS (USICT Student Dataset) using Network Science, Newman-Girvan Algorithm.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124813014","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974518
Preeti A. Kale, M. Nene
In Wireless Sensor Networks (WSNs), Data Aggregation Trees (DATs) are employed for energy efficient data gathering. Energy efficient data collection is a primary requirement in the smart world of Internet of Things (IoT) as it facilitates to extend the survivability of the network. DATs gather data efficiently by employing data aggregation functions at the aggregator nodes. The employed aggregation function influences the cost of communication and cost of computation at a node. The study in this paper presents the techniques to estimate the communication and computation costs incurred for DAT construction, asymptotically. The strength of the proposed techniques is its ability to enable the estimation of best, average and worst case cost of DAT construction and rescheduling scenarios. Based on the asymptotic analysis, the study in this paper demonstrates the utilization of the proposed techniques to estimate the best and worst cases for communication and computation cost to meet the design objective of adhoc WSN deployments.
{"title":"Asymptotic Cost Estimation for Scheduling Data Aggregation Trees in Sensor Networks","authors":"Preeti A. Kale, M. Nene","doi":"10.1109/ICCCIS48478.2019.8974518","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974518","url":null,"abstract":"In Wireless Sensor Networks (WSNs), Data Aggregation Trees (DATs) are employed for energy efficient data gathering. Energy efficient data collection is a primary requirement in the smart world of Internet of Things (IoT) as it facilitates to extend the survivability of the network. DATs gather data efficiently by employing data aggregation functions at the aggregator nodes. The employed aggregation function influences the cost of communication and cost of computation at a node. The study in this paper presents the techniques to estimate the communication and computation costs incurred for DAT construction, asymptotically. The strength of the proposed techniques is its ability to enable the estimation of best, average and worst case cost of DAT construction and rescheduling scenarios. Based on the asymptotic analysis, the study in this paper demonstrates the utilization of the proposed techniques to estimate the best and worst cases for communication and computation cost to meet the design objective of adhoc WSN deployments.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121756145","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974468
Shubham Mishra, Zeesha Mishra, B. Acharya
A high-speed cryptography algorithm is the necessity of today’s world to communicate between two resource-constrained tools. We have proposed a 128-bit block size and 256-bit key size LEA cipher for high speed and throughput purpose. High speed is an essential requirement with the secure transmission of information. Pipelining is the best way to achieve these requirements with higher throughput. The implementation results have shown that this method has a higher capability of speed and throughput as compared to previous ciphers. LEA has security against all existing cipher attacks and compatible with both hardware and software. Proposed work is 73.7% and 62.85% improved version of compared paper for respectively throughput and speed.
{"title":"A High Throughput And Speed Architecture of Lightweight Cipher LEA","authors":"Shubham Mishra, Zeesha Mishra, B. Acharya","doi":"10.1109/ICCCIS48478.2019.8974468","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974468","url":null,"abstract":"A high-speed cryptography algorithm is the necessity of today’s world to communicate between two resource-constrained tools. We have proposed a 128-bit block size and 256-bit key size LEA cipher for high speed and throughput purpose. High speed is an essential requirement with the secure transmission of information. Pipelining is the best way to achieve these requirements with higher throughput. The implementation results have shown that this method has a higher capability of speed and throughput as compared to previous ciphers. LEA has security against all existing cipher attacks and compatible with both hardware and software. Proposed work is 73.7% and 62.85% improved version of compared paper for respectively throughput and speed.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"180 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121250451","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974490
Manjeet Singh
Grid Computing is basically an infrastructure which provides high computational capacity to the distributed system by making use of widely geographically distributed resources. The resources in grid are owned by different organizations which has their own policies, computation capability, framework, and cost and access model. The last fifteen years have observed a growth in computer and network performance due to better hardware and software availability. Although, the growth in field of computing is very good but still there are problems in the field of Science, Engineering and Businesses which needs to be handle more effectively. Actually due to the large size of the problems, numerical computation, large volume of data processing, the problems needs a verity of resources which are often not available at a single machine. So, basically Grid Computing is an idea to identify and use the geographically distributed heterogeneous resources to solve the data intensive numerical problems. There are various approaches which make use of distributed resources in different manner to solve similar kind of problems known by different names like-Cluster Computing, Cloud Computing, Meta Computing, Scalable Computing, Distributed Computing, Parallel Computing and more recently the grid computing.
{"title":"An Overview of Grid Computing","authors":"Manjeet Singh","doi":"10.1109/ICCCIS48478.2019.8974490","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974490","url":null,"abstract":"Grid Computing is basically an infrastructure which provides high computational capacity to the distributed system by making use of widely geographically distributed resources. The resources in grid are owned by different organizations which has their own policies, computation capability, framework, and cost and access model. The last fifteen years have observed a growth in computer and network performance due to better hardware and software availability. Although, the growth in field of computing is very good but still there are problems in the field of Science, Engineering and Businesses which needs to be handle more effectively. Actually due to the large size of the problems, numerical computation, large volume of data processing, the problems needs a verity of resources which are often not available at a single machine. So, basically Grid Computing is an idea to identify and use the geographically distributed heterogeneous resources to solve the data intensive numerical problems. There are various approaches which make use of distributed resources in different manner to solve similar kind of problems known by different names like-Cluster Computing, Cloud Computing, Meta Computing, Scalable Computing, Distributed Computing, Parallel Computing and more recently the grid computing.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126673294","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974464
Pavan Kumar Pandey, A. Swaroop, Vineet Kansal
In VANETs, the vehicles communicate with each other either directly or using some infrastructure. Due to high mobility behavior of nodes in VANETs, communication among vehicles faces several challenges. Additionally, there are other factors also which effect routing performance in VANETs such as vehicular density, link stability, and vehicle speed etc. Several efficient routing protocols have been designed to route messages between nodes. Moreover, different routing approaches can be used based on the application scenario and network situation. Therefore, selection of appropriate routing strategy is an important issue in VANETs. The present paper discusses the most recent routing protocols proposed for VANETs in last five years. The discussed protocols have been divided in to three categories namely, geographical or position based, topology based and cluster based routing. The protocols have been analyzed for their advantages, disadvantages and applicability in various scenarios.
{"title":"A Concise Survey on Recent Routing Protocols for Vehicular Ad hoc Networks (VANETs)","authors":"Pavan Kumar Pandey, A. Swaroop, Vineet Kansal","doi":"10.1109/ICCCIS48478.2019.8974464","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974464","url":null,"abstract":"In VANETs, the vehicles communicate with each other either directly or using some infrastructure. Due to high mobility behavior of nodes in VANETs, communication among vehicles faces several challenges. Additionally, there are other factors also which effect routing performance in VANETs such as vehicular density, link stability, and vehicle speed etc. Several efficient routing protocols have been designed to route messages between nodes. Moreover, different routing approaches can be used based on the application scenario and network situation. Therefore, selection of appropriate routing strategy is an important issue in VANETs. The present paper discusses the most recent routing protocols proposed for VANETs in last five years. The discussed protocols have been divided in to three categories namely, geographical or position based, topology based and cluster based routing. The protocols have been analyzed for their advantages, disadvantages and applicability in various scenarios.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114732649","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 : 2019-10-01DOI: 10.1109/ICCCIS48478.2019.8974495
J. Kaur, J. Singh
For processing any natural language processing application, the knowledge of structure of sentence including its boundaries plays a vital role. Incorrect sentence boundary may lead to wrong outputs and hence decreasing the performance of NLP systems. Detecting sentence boundaries in code mixed social media text is not an easy task. People generally omits the boundary markers and use punctuation for other stylistic tasks. We propose a deep neural network approach for sentence boundary marking as well as suggesting appropriate punctuation mark in code mixed social media text. We experimented with single layer bidirectional and two layer bidirectional models. Both word sequence and character sequence are experimented. Bidirectional model using character sequence out performs all other models for sentence boundary detection as well as end marker suggestion.
{"title":"Deep Neural Network Based Sentence Boundary Detection and End Marker Suggestion for Social Media Text","authors":"J. Kaur, J. Singh","doi":"10.1109/ICCCIS48478.2019.8974495","DOIUrl":"https://doi.org/10.1109/ICCCIS48478.2019.8974495","url":null,"abstract":"For processing any natural language processing application, the knowledge of structure of sentence including its boundaries plays a vital role. Incorrect sentence boundary may lead to wrong outputs and hence decreasing the performance of NLP systems. Detecting sentence boundaries in code mixed social media text is not an easy task. People generally omits the boundary markers and use punctuation for other stylistic tasks. We propose a deep neural network approach for sentence boundary marking as well as suggesting appropriate punctuation mark in code mixed social media text. We experimented with single layer bidirectional and two layer bidirectional models. Both word sequence and character sequence are experimented. Bidirectional model using character sequence out performs all other models for sentence boundary detection as well as end marker suggestion.","PeriodicalId":436154,"journal":{"name":"2019 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128614417","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：