Online Social Networks - Science topic

Academic social networks such as ResearchGate (RG) and Academia.edu can be valuable resources for implementing under and postgraduate learning activities. These platforms provide access to a vast amount of research papers, scholarly articles, and other academic resources that can supplement traditional coursework and reading materials. Students can use these platforms to explore topics in greater depth, find additional sources for their research papers or projects, and connect with other scholars and experts in their field.

It is important to note that academic social networks should be used as a supplement to traditional learning resources, rather than a replacement. While these platforms can provide valuable resources and connections, they do not offer the same level of structure and guidance that is typically provided in a classroom or through traditional coursework. As such, students and faculty members should use these platforms in conjunction with other learning resources to ensure that they are getting a well-rounded education. Additionally, it is important to use caution when using these platforms, as not all of the content posted on them is peer-reviewed or reliable.

linked in

Any journal indexed in scopus

Dear Samuel Akwasi Frimpong ,

Look the link, maybe useful.

Regards,

Shafagat

of course, as soon as we publish it (I hope in less than a month!).

I am interested in working on online social networks and social networking analysis

Hi.,

The rapidly popularized online social networks (OSNs) greatly enhance the speed and ... Therefore, mitigating the impact of bogus rumors is of primary importance.

For More info:

That's great Jaqueline Brinn ! Thank you very much!

Hierarchy and size are two different independent aspects.

Also hierarchy can not be labelled as strong or weak......

Thank you very much for your answer George! However, I wasn't meaning Digital/Web atmospherics but actual physical atmospherics. Like a service or restaurant design or a brand experience physical/atmospherical design.

Dear Mahdieh Zabihimayvan ,

Overall, your model did a good job capturing the edge term (first GOF evaluation). Your MCMC diagnostic depicts a model that does not degenerate which is a good news. For the rest of the structure of the network, your model somewhat performs poorly, failing to correctly represent the structure of the observed network. This means that there is still room for improvement regarding how you specify your model.

Depending on the size of the observed network, it is usually difficult to fit a model that correctly reproduce the structure of the observed network. The usual strategy (which I would recommend here) is to fit various model specifications, then choose the best based on the highest log-likelihood (or alternatively the lowest AIC or BIC), and evaluate it using the GOF (including an ROC curve). What you should avoid at all cost is a model that has a good fit but failed the MCMC diagnostic assessment.

I hope this helps!

Thank you,

Roseric

I second NodeXL as well. You can also look into TAGS- https://tags.hawksey.info/

Dear Faisal,

If you want to get data for a specific geographical area, you can use Gephi with "Twitter Streaming Importer" plugin.

Best,

İbrahim

"The followers of the users have similar follower count as the user himself" I think this is an interesting finding which is worth looking up on. If the claim is true then you can look at possibility of how this affects the data analytics in social network. For example if the network is using collaborative filtering approach to recommend users or tweets then whether or not this fact is used to calculate the similarity score.

Thank you Ivan, I'll visit these sites and tell you my impression.

Thank you for your email.

With all my best,

Samah.

Dear Ja'far,

Have you found the data set you were looking for? Because I am looking for the same data and I would be grateful if you let me know what data set you found.

I find this paper :

A Survey on Information Diffusion in Online Social

Networks: Models and Methods

In case you have accounts of public figures, you could use crowd sourcing platforms and ask users to verify or indicate the mistakes in the system output. However, if the accounts belong to random users, I would advise against that due to privacy concerns, even if the accounts are public.

Kevin, here are two sources but they will lead you to many other sources:

If you have other questions or would like other resources, I will dig into my own library.

MHC

I think it depends on what you are trying to measure.  Are you simply trying to measure whether a member of the general public can see information posted by an experimental subject?  Are you trying to measure how much data posted by a participant is used by Facebook for marketing purposes?  Are you trying to measure the participant's expectations for privacy when making self-disclosure online?  

The instrument you choose must be based on which will answer your research questions.

you need to look for a measurment study paper of facebook. There you will find values you are looking for. For example, avg degree, cluster coefficient, diameter and separation between nodes (shortest path). These values can be used to generate a new graph to mimic facebook.

start from here,https://research.facebook.com/blog/three-and-a-half-degrees-of-separation/ you will find the shortest path and graph diameter of facebook. you need average degree, https://www.facebook.com/notes/facebook-data-science/anatomy-of-facebook/10150388519243859/. 100 freinds and so on.

It is easy to find the rest  

I agree to explanations provided by Henrique Santos and Jean Damascene Twizeyimana. Info system encompasses users so, models developed for Info System Security applicable to humans also. and yes, if as a human being I will not take the responsibility of safe keeping of my gadgets (s.a, my cards, smart phones, physical keys, laptop etc) or will not  keep certain things private to myself  (e.g., my PINs, Password, OTPs etc) then I doubt anything can provide me security.

Hello,

Hope these cane guide you:

 https://www.researchgate.net/file.PostFileLoader.html?id...assetKey...

If you want something user friendly, hootsuite offers quite a lot of options

With the exception of the developing world, just about every young adult carries a cell phone. Every cell phone has a stopwatch. You could ask participants to use their stopwatch to monitor screen time and then have them text the length the stopwatch ran at the end of each day.

Many of these are open API if you partner with someone who codes you can just query for what you need. Similarly, Nvivo has a plug in for downloading data/scraping data right from Facebook/Twitter.

Thank you Mr. Aladwani and Mrs. Mierzecka for your answers! I will look into the references. They appear to be something different from I am looking for, since I am interested in a trait variable, such as an approach/avoidance predisposition or intravert/extrovert, but in social networks. 

I agree with Yulia: unless you have strong reasons to prefer data from a company (for example, access to historical data), I would definitely build my own dataset. It is not extremely hard to do it, specially if you have some computational skills. But even if you don't, there are ways to circumvent it and get your data. Should you have more questions, I'll be happy to help. 

I think a FB survey can be reported similarly to other SNS investigations.  There is an e-book 'Otto, A. (2011). Does Facebook make you happy?', and he used similar ways to report response rates, sampling etc.

"Sample

The sample employed for this research project consists of international young adults using Facebook. We focused on young adults (18 - 29), because they seem to be the main users of Facebook. 291 participants started the main part of the survey, but only 207 participants completed it. Of the 207 participants, 84 completed both the main survey and the follow-up survey. The respondents mean age is 24.19 (SD = 2.38) with 72 males (34.8%) and 135 females (65.2%). 55.6 per cent (115) of the participants answered “studying” to the question of their main current occupation, 48.6 per cent (80) “working” and 5.8 per cent (12) were unemployed at the moment participating the survey. The nationality of the respondents shows a high variety, with the largest group coming from Europe (82.1%)."

I think that as long as you can describe the usual details of who the survey was distributed to, what criteria were used for inclusion, how many responded etc, then it should be quite straight forward. If the data is not available to use the same statistical analysis you have used in the other methods, then opt for a simple descriptive approach.

Dear Mennatallah,

good day,

Kindly find the attached file regarding one of my researches, it may help you

best regards,

Dr.Alhaddad

Do you know the IAT (Internet Addiction Test, do not mix it with Implicit Association Test)? it may helps you if you decide to create one specific for your research

Hi Mona,

Co-founder of Prolific Academic here. We've developed an alternative to MTurk tailored for academic research and we are accessible worldwide. Over 200+ researchers have completed over 750+ studies on our platform so far.  

New users can run a trial study for free: https://www.prolific.ac/rr?ref=5ZFZ276D. You can find more information here: www.prolific.ac

Should you have any questions about the platform please don’t hesitate to get in touch.

Cheers,

Katia

The Netvizz tool is very useful. It also has an accompanying academic paper, see link.

One of them is Dunbar, R. I. M. 

Dear Rasha:

Wonderful topic of discussion! Rumor is a potentially harmful social phenomenon that has been observed in all human societies in all times. Social networking sites provide a platform for the rapid interchange of information and hence, for the rapid dissemination of unsubstantiated claims that are potentially harmful.

In a recent research study (Tripathy, Bagchi,  & Mehta, 2013) argued different methods for combating rumors in social networks actuated by the realization that authoritarian methods for fighting rumor have largely failed. Authors major insight is that in situations where populations do not answer to the same authority, it is the trust that individuals place in their friends that must be leveraged to fight rumor. In other words, rumor is best combated by something which acts like itself, a message which spreads from one individual to another. The authors call such messages anti-rumors. The authors studied three natural anti-rumor processes to counter the rumor and present mean field equations that characterize the system. Several metrics are proposed to capture the properties of rumor and anti-rumor processes. The metrics are geared to capture temporal evolution as well as global properties of the processes. The authors also evaluated  methods by simulating rumor and anti-rumor processes on a large data set of around 10^5 nodes derived from the social networking site Twitter and on a synthetic network of the same size generated according to the Barab\'asi-Albert model.

Another recent study (Zhao, Lin,& Guo, 2014) offered a new method for company managers to predict the variation trend of demand and estimate demand loss when a firm is attacked by rumors!

Finally in a fresh research study (Wang, Vasilakos, Ma,  Xiong, 2015) explored the pattern of gossip diffusion in social networks when uncertainty exists in users’ decision making. The obtained results perfectly comply with the philosophical saying about rumor diffusion in real social life: easy come, easy go!

Hope this helps!

Nadeem

A bit off-topic perhaps, but in my courses I've decided to leave Facebook for the students. The classes I teach all have Facebook groups where they help each other, the groups use Facebook for their assignments. While I would love to have access to these groups for research purposes, I've found that the students prefer to be left alone in these groups and do their own thing without teacher intervention, so I've chosen to respect that.

Some results here between hashtags and viral reach?

Hi;

Maybe you can get some help out from our attached article.

Best,

Erik Amnå

Hi Prem

I know two good methods of ranking. The first of them is Pairwise comparison method published by Waldemar Koczkodaj articles "Testing accuracy Pairwise comparison" and "On Axiomatization of Inconsistency Indicators for Pairwise Comparisons"  . The secons is Objective ranking published by Andrzej P. Wierzbicki "Reference Point Approaches and Objective Ranking"

Jan Grzegorek

Hi Muhammad,

Mmmh, you will be surprised, that the "Social Network Theory" itself might be a good starting point. Literature dealing/using  this theory, plus their empriical conrtibutions might help you navigate well around various SNS related studies, etc.

Some good articles and text books references are:

Burkhardt, M.E. (1994). Social interaction effects following a technological change: a longitudinal investigation. Academy of Management Journal, 37, 869-898.

Freeman, L. C. (1979). Centrality in Social Networks: Conceptual clarification. Social Networks,1, 215-239.

Freeman, L.C., White, D.R., & Romney, A.K. (1992). Research methods in social network analysis. New Brunswick, NJ.: Transaction Publishers.

M.S. Granovetter., "The strength of weak ties: A network theory revisited," Social Structure and Network Analysis (P.V. Marsden and N. Lin, Eds.). Sage, Beverly Hills CA, 1982, pp. 105-130.

Ibarra, H., & Andrews, S. B. (1993). Power, social influence, and sense making: Effects of network centrality and proximity on employee perceptions. Administrative Science Quarterly, 38, 277-303.

Meyer, G.W. (1994). Social information processing and social networks: A test of social influence mechanisms. Human Relations, 47, 1013-1048.

هل يمكن ربط الاداره المركزيه للارشاد الزراعى بمعهد بحوث الارشاد الزراعى ومعاهد المركز ربط الكترونى لتفعيل دور الادارة المركزيه للارشاد وقطاع الارشاد الزراعى بالوزاره.....ماهى اهم مقترحاتك؟

In this research they talked about "passive forms of communication" in Facebook https://www.facebook.com/note.php?note_id=55257228858 and I published a summary (in Spanish) here http://personal.us.es/isidromj/php/2009/06/haciendo-amigos-en-facebook/

Regards,

Isidro

InsyAllah I will explain in detail soon.

Betweenness is very important for "brokers". No matter which metric you decide to use in the end (including defining your own), you will probably need to get betweenness results as well and include them in your analysis.

India is the second largest bamboo producer in the world. After agriculture, bamboo is a major sector that generates income and employment especially in the remote, rural and backward regions of the country. For the north eastern region, bamboo is an even more important and vital resource. The bamboo stock of the north eastern region constitutes nearly 66% of the nation’s bamboo assets. Bamboo is a unique and highly renewable resource with multiple uses. It is the best natural engineering material, excellent for soil management, energy efficient and excellent in carbon sequestration. It provides direct benefits in income and employment in rural and backward region at all levels ranging from plantation, artisan use, primary processing and downstream industrial activities. More than 70% traditional housing uses bamboo as an important structural material.

You might want to look at this work by Yong et al

http://www.kmel-journal.org/ojs/index.php/online-publication/article/viewFile/148/121 

This one also just popped up on my feed: 

Hi Nima, what you are probably looking for is the term "Science 2.0". Science 2.0 bascially tackles the investigation of how Web 2.0 changes scholarly communication, collaboration and cooperation among researchers, then research and publication processes in general (think of e.g. the Open Access movements or concepts like Open Peer Reviews and Altmetrics). Science 2.0 is a pretty new field of interest and while it may lack a clear definition at this point, there is a large body of exiting research going  into this direction, for instance check out the background paper the European Commission provides:

Or e.g.

RG can be improved in following ways:

1-by adding the possibility of automatic multiple downloads (like science direct)

2- Removing the downvote button or at least prohibiting multiple downloads in the same question by one person.

3- checking and Removing the duplicate (or very similar) questions by editors.

You can use facebook's graph API, for example. Find some code here: http://smallstats.blogspot.de/2014/04/some-fun-and-insights-with-facbooks.html

Evolution of social networks:

Addiction to any social networks will definitely reduce the productivity.

After Google+, RG?

My pleasure. I should have perhaps added that the only bad feature of RG is the anonymous down vote. This is a real negative and gives opportunities for underhand dismissals of member contributions that are of no practical benefit and turn scientific debate into a cheap talent show. Anyone who want to disagree with fellow researchers should have the courage and decency to identify themselves, put up a counter argument or shut up and keep out of the debate.

Hi

please check this link: http://sloanreview.mit.edu/collections/creating-value-with-social-business/.

There are number of crawlers available like:

1. Crawler4j: is an open source Java crawler which provides a simple interface for crawling the Web.

2. Apache Nutch: is a highly extensible and scalable open source web crawler software project.

3. Ex-Crawler: is divided into three subprojects. Ex-Crawler server daemon is a highly configurable, flexible (Web-) Crawler, including distributed grid / volunteer computing features written in Java. Crawled informations are stored in MySQL, MSSQL or PostgreSQL database.

4. Whalebot: is open-source web crawler. It is intended to be simple, fast and memory efficient. It was created as a targeted spider, but you may use it as common.

You can take the help by looking at the source code available on the respective websites and design your crawler accordingly.

In case of any more query then feel free to contact.

With the explosion in information technologies, information system researchers had conducted data concerning users’ intentions for adoption of the new technologies, most of which based on Davis’s (1989) Technology Acceptance Model (TAM), which becomes the mostly applicable and useful model for computer use. The TAM is the information system theory of modeling how users come to accept and use a technology. The theory states that there are several factors can influence the user decision about how and when he will use a new technology.

Moreover, TAM theory lies in a number of psychological theories. One of which is the Theory of Reason Action (TRA). TRA is a behavioral well-known model which had been extensively used in information system research. This theory states that reasoning flows from belief and evaluation to the development of attitude towards performing behavior.

I would say I completely agree with you due to two facts. First, the membership of social networking sites has become a necessity rather than an option. Lastly, as a result of the huge amount of social networking sites' users participation and interaction, this can be distinctly observed.

At a practical level, when starting, you can use facebook posts as a class exercise (if its relevant to topic obviously :o) to help think through how to thematicize content. I do this as its a good way to get business students talking about controversial social issues relating to business problems. Start with a vigorous debate of interest. Print out all of the comments, cut them up with scissors and then get students to organize into themes. Usually posts responding to a controversial issue are polarised, so for and against is a good place to start, and then start to tease out the issues within each post and look for patterns. This is a good practical way to get started with a research project because the discussions with students then help the researcher realise all the fishhooks as well in thematicising content (e.g. hypertext linking behaviours; mini outbreaks of warfare between posters; chronological changes in the tone of posing as debate settles down etc..).

Another useful method is in conversational analysis and mapping techniques.

The other way is simply to use exel and do it the old-fashioned way.

I see that software like leximancer can be used for this as well, to generate concept maps but there needs to be a lot of data and the conceptual work still needs to be done.

Not sure whether this will help you, see in the middle of the page "download".

Dear Nima,

I tried to set up some social network in science and I found :

- Facebook works for students

- Twitter works good for journal, researchers, institute

- Google+ does't work too much, but it seems that it is the future….

Concerning social network specialized in science, I didn't use yet. I will.

I think social network is really important to communicate : a table with two chairs stay the best but we live in a strange world with we don't have time anymore for that (we collaborate with people from all other the world, so it's time consuming to take a plane each time you have a question).

As a starting place you might take a look at Jan Boehmer's work on the relationship between liking a link on Facebook and browsing behavior on the linked website. It isn't quite what you are looking for, but his citations might be a good starting place.

In addition (I haven't read them):

Regards,

Joachim

You can also extract Twitter data with NodeXL. It's a plug-in for MS Excel and - from what I've heard since I haven't used it myself - seems to be quite user-friendly: http://nodexl.codeplex.com/

@Hansika: As to your question... Yes, I think that could be a way to assess is. Normally, the uploading and sharing isn't a problem due to the wider and wider spread of the Internet and people's adoption of smartphones. So, it could be an interesting scenario to observe people when they are not able to share, e.g. in the context of an experiment or ethnographic study. Does this result in stress or negative feelings? Does it lead to dissatisfaction? One could measure this with questionnaires or tracking devices.

@Hansika & Dian: Very interesting discussion emerging here!

The process you described, Dian, is a plausible one in many regards and it probably happens that way to some extent. The question then is: How can we investigate it empirically? How widespread and prevalent is it? As for the first question, it's probably very hard to study such a complex process in a proper way and to isolate the behavioral change associated with the process. In a first step, it could be a good idea to explore the field with a qualitative study.

As for the second question, there are a number of confounding factors that make it hard to estimate the prevalence of the process. First, studies have shown that people are very bad at evaluating their audience on SNS because of invisible audiences and algorithmic filtering (which many users are not even aware of). Both factors make inferences about audiences difficult (http://dl.acm.org/citation.cfm?id=2470658=). Second, even if users have a certain audience in mind and provide a self-image that tries to satisfy this imagined audience (which diverges sharply from the real audience), the process is maybe more incorporated, emotional, sub-conscious or habitual than cognitive. Third, (imagined) audiences are in many cases quite diverse: Friends, family, colleagues and acquantainces might all have different opinions on what's a likeable and cool status update. How can I, as user, provide an image that satisfies the diverse "stakeholders" best?

These points make it difficult to tell how much - and how many - people change their behavior because of self-monitoring on SNS.

I agree with you both that the small town analogy captures many aspects of what's going on on some SNS. Going back to the German sociologist Tönnies we could argue that Facebook and similar SNS resemble "Gemeinschaft"/community more than "Gesellschaft"/society (https://www.researchgate.net/post/What_is_the_meaning_of_terms_Gemeinschaft_and_Gesellschaft).

BTW, there is a discussion on a similar topic going on in another thread that might be interesting for some of you. Feel free to check it out here ;) https://www.researchgate.net/post/Do_social_media_make_us_less_individual

Don't forget other applications like Gephi (https://gephi.org/), R (for example the sna or statnet packaga; http://www.stanford.edu/~messing/RforSNA.html), ORA (http://www.casos.cs.cmu.edu/projects/ora/) or Ucinet (http://www.analytictech.com/ucinet/)! All of them are great and offer tons of cool applications.

Online Social Networks and Recruitment are very potent subjects for research, congratulations on your thesis subject. Although exciting this topic is also complex.

I think many recruiters and job seekers would be glad to know we have reached that time where you could screen a candidate's ability for a certain role by analysing his activity on online social networks (James' answer did a very good job on shortlisting all the major SN's, props for including startups too !). What people tend to omit very often and to much surprise is that the vast majority of online personas are simply a fabrication of the user and his network. This is certainly not an indicator of performance neither past or present, let alone future.

For now the landscape is fragmented - with multiple environments, each with their own dynamics. While in some industries (IT being the foremost) there are emerging concepts such as freelance workers portals even those cannot be considered good predictors of future job performance.

My opinion is that with new initiatives, see for example gamification, which are embedded in day to day working processes we will have (real) data for performance tracking and predicition. For an example just look at this website - Research Gate. While I am typing an answer they are suggesting this could increase my RG Score.

The ultimate social network for recruiters would allow you to browse operations, which are specific to the company's operational profile and actually undergo tasks which either have a direct and measurable impact on the recruiting organization or are designed through high-fidelity simulations to replicate real processes. I think that would definitely be something a lot of recruiters and job candidates would be excited about.

Researchersˊ information seeking behaviour are activities that researchers do it for accessing their required Information.

Maybe could you be interested: e-chupal is an empirical Social network patrocinated by ITC in soybean cultives

Dear Diane, I think it very much depends on your country's legislation and your IRB. Where I am if a blog is public its data can be analysed without any consent needed by the author, it's public domain. Our IRB usually waive the need for informed consent forms in this case. As an additional precaution and respect for the blogger's privacy, before I publish any data I check to see if the blog was made private or was cancelled by the blogger. If so, I delete its data from my database and rework my analysis.

If possible, I also try to keep out any identifying information from my results and I do not divulge direct link to the blog as to protect the blogger's privacy.

If you can contact the authors, I would certainly try to get their permission though.

Do you already have the profiles downloaded? If you do, then you could just number them all and randomly select a subset, using something like this: http://www.random.org/integer-sets/

There are SSCI indexed quality journals like Computers in Human Behavior. Here are some sample articles you may be interested.

Dohyun Ahn, Dong-Hee Shin. Is the social use of media for seeking connectedness or for avoiding social isolation? Mechanisms underlying media use and subjective well-being. Computers in Human Behavior, Volume 29, Issue 6, November 2013, Pages 2453-2462

A. Contarello, M. Sarrica. ICTs, social thinking and subjective well-being – The internet and its representations in everyday life. Computers in Human Behavior, Volume 23, Issue 2, March 2007, Pages 1016-1032

Joanna Mitchell, Rosanna Stanimirovic, Britt Klein, Dianne Vella-Brodrick. A randomised controlled trial of a self-guided internet intervention promoting well-being. Computers in Human Behavior, Volume 25, Issue 3, May 2009, Pages 749-760

Scott Caplan, Dmitri Williams, Nick Yee. Problematic Internet use and psychosocial well-being among MMO players. Computers in Human Behavior, Volume 25, Issue 6, November 2009, Pages 1312-1319

Show preview | PDF (199 K) | Related articles | Related reference work articles

Ssu-Kuang Chen. Internet use and psychological well-being among college students: A latent profile approach. Computers in Human Behavior, Volume 28, Issue 6, November 2012, Pages 2219-2226

There is a nice compilation of SNA tools here: http://en.wikipedia.org/wiki/Social_network_analysis_software

Just search for "browser" - to get the online tools ( I think there are 4 mentioned in the list)

If members of one team overlap with another, then an issue of "double counting" might be present in the regression estimates. While other models such as Naïve Bayes are not sensitive to violations of the independence assumption (and have shown to overpass accuracy of linear regression estimates), regression models highly depend on the normality of the error distribution of the estimates, thus are sensitive to such violations of the independence assumption. There might be a possible way to bypass this problem by doing a weighted regression by discounting (inverse weighting) by the number of overlaps. For example, if a member does belong to a single group, the regression weight is 1.0, for a member belonging to two groups, the regression weight is 0.5, and for a member belonging to all three groups the regression weight is 0.333. This is my two-pennies contribution to the discussion.

Critically the answer depends on your question and your definitions of that form the basis of your network. Every metric is simply defined within a mathematical formulation bounded by assumptions. So you need to understand what centrality means to YOU very carefully and whether it makes sense to apply it across the network so you reach a single value (especially given the largest component only is used normally). Another approach might be to select each cluster and calculate the metric with a view to understanding the distribution of the metric for your network. Carter Butts (2009) in the Science article titled 'Revisiting the foundations of network analysis' outlines how careful we need to be to understand the exact nature of the linkages before we can proceed.