Summary by Patrick This paper studies how users communicate and build ego networks on Twitter in hopes to futher understand the dynamic social structures of humans' online. Users are divided into three categories based on their active lifespan. These are divided by the number of days the user has been active. By constructing ego networks based on temporal windows, the authors hope to observe how the ego networks evolve over time. This is done by studying timeslices of one year from when the user joined and by classifying the users and analysing their history. Results show that users on Twitter do not follow the same social patterns as they do on other social networks. The average user maintains a smaller ego network with weaker ties and a high turnover. The authors appear to have gained a lot of data before starting this analysis, but the number of profiles crawled appear to be a small part of the number of users at the time of the analysis. Summary by Zahid Dynamics of Personal Social Relationships in Online Social Networks: a Study on Twitter Valerio Arnaboldi, Marco Conti, Andrea Passarella Robin Dunbar Summary Now a days the online social networks are getting very important means of communication. Through this we can get very closed to our families and friends. Sharing the knowledge and ideas from online social networks is a useful way. Online Social Networks are getting more popularity because of it ability to transform people into active procedure of information, letting them create, access and share contents from anywhere. OSN are producing strong effects on our society and they are changing the behavior of society but few areas where they are impacting on human social behavior is still unknown. In the future these networks will be of primary importance in communications. So it is very important that we should study the human behavior because it impact on these social networks, it will help us into, how users can contribute to the process, designing future OSN so we can manage the social relationships through digital communications. In this paper writers analyze the Twitter data to set containing communication traces of more than two millions users to study the dynamic properties of the behavior of OSN users. There are two types of social networks one of them is ONLINE and second one is OFFLINE. Authors of this paper obtained the last 3200 tweets of a large data set of users for analysis about the evolution of human social behavior in Twitter over time. According to the analysis of authors that people prefer to maintain weak social relationships than strong ones, with a high turnover of contacts in their networks. There are main two types of users those can be divide like, users who have a short , but intense, activity and users who interact with social peers for ling time intervals. They used one concept that is call ¡°ego network¡±, its mean that analyze at the local properties of personal social networks. Ego Networks describes the social relationships between an individual, called ego, and all the contacts ego has with other people, called alters. According to Robin Dunbar there is a limit on the number of alters people can actively maintain in their networks, due to the cognitive constraint of human brains in ego networks. There are four layers in ego networks and each of them have different number of members. Figure 1 shows the ego network layers. ? First layer is called support clique having 5 number of members, second one is called Sympathy layer and it has 15 members. This one layer is called by affinity group and it has 50 members, and fourth one is active network having 150 members in it. Authors got the data through Twitter ( it is an online social network service that is very important and having high popularity with more than 500 millions registered users as of 2012) Results show that ¡°Specifically, the active network has a total decrease of 30.73%, the sympathy group of 45.91% and the support clique of 53.22%. Regular users show a different behavior, with a considerable increase in the active network size (31.16% in almost 4 years), but with a decrease in the other layers (32.17% for the sympathy group and 30.42% for the support clique).¡± In a nutshell this paper presents a detailed analysis of the dynamic processes of Ego Networks and personal social relationships in Twitter. So with the help of results this configured that human behavior in Twitter significantly differs from other social networks. Through another point of view the results also indicate that users do not immediately abandon Twitter tend to use. It shows that the hypothesized decline in the use of Online Social Networks might not be present, at least in Twitter. Summary by guangyu ----- BEGIN SUMMARY ------ Dynamics of Personal Social Relationships in Online Social Networks: a Study on Twitter ------------------------------ This paper analyses a data set of Twitter communication records to study dynamic processes that govern the maintenance of online social relationships. In the section 2 it introuce us the similarity study in offline social networks which is really interesting .and also introduce the previous study on this eara. The Dunbar number is proved as 150 which is bigger than that in twiter. In section 3 it descibe how to collected data and analysis .The decting hunman mechanism is referred from previous study. and a remarked phenomenon is only 1/4 data collected are selected . In the section4 3 different groups were given according to the fluency of contacting and also have a classification of user by their lifespan. another innovative study is analysing the recency of contact. after giving results in section5 ,the paper ends with discussion about the differences between and traditional social networks. Suggested questions for discussion: the differences structure (clustering coefficient, degree distribution, Average path length)betwwen twiter and other social network and the reason