Consensus Building in Collaboration Networks

Main impacts

  • Successfully identified trusted users in online Q&A networks
  • Identified factors that contribute to consensus building

In our informal learning settings, while seeking for online help, we interact with our peers, define and discuss certain problems, exchange opinions and try to reach some kind of consensus on a best provided solution. Such interactions between a help-seeker and more experienced peers are also characterized as scaffolding in networked online environments. Our focus here is to address the scaffolding phenomenon in online settings by studying the role of learners’ social status on negotiation processes. Negotiation processes include the involvement of learners on understanding and defining certain problems, identifying the experts in the field and finding possible solutions. Our research in terms of the Layers project has shown that the presence of high social status learners, known as hubs, in a networked online environment is crucial for negotiation processes, because such learners can distribute a single common opinion to a large number of other learners. Additionally, the consensus building process is indeed affected by the learners’ social status. But, there is a specific setting for each collaboration network, in which social status speeds up the consensus building process.

The situation before Layers

Highlights

In online scaffolding settings, we define problems and try to solve them with the help of the online community. The best solutions are usually found while interacting with many individuals that we do not know personally. It is difficult to identify the social status or the working experience of the individuals with whom we interact, if these are not made explicit by the system. This hinders users seeking for help to trust the online community because they can not relate the quality of the provided answers or solutions to the level of expertise or social status of individuals that proposed those solutions. Within the Layers projects, [1] and [2] conducted two studies, one in the largest Estonian online forum for construction professionals, and second in UK and German professional forums. These studies discovered that the identification of the more authoritative opinions about a problem solution is one of the key issues.

The role of users’ social status on negotiation processes taking place in a networked learning environment was not tackled by state of the art research yet. Within the Layers project we investigated the influence of social status on negotiation processes that take place in collaboration networks.

Description

In our informal learning settings, while seeking for help online, we interact with our peers, define and discuss certain problems, exchange opinions and try to reach some kind of consensus on a best provided solution. Such interactions between a help-seeker and more experienced peers are also characterized as scaffolding in networked online environments [1]. It is our natural predisposition to interact with individuals who have a high social status in our community. Customarily, our social interactions and, to some extent, our behavior are influenced by actions of individuals with a high social status. However, state of the art research does not provide facts on the influence of social factors on the negotiation processes taking place in a networked online environment. Negotiation processes include the involvement of learners on understanding and defining certain problems, identifying the experts in the field and finding possible solutions. As such, negotiation processes affect the professional knowledge accumulated in the community as a whole [1].

In terms of social factors or social context, we consider the social identity of learners within a community, which is reflected through their social reputation or status that they earned over their lifetime. Of course, these factors cannot be seen as isolated so they are related to: (i) trust that they earned over a period of time or convincing power, which in turn are gained through the learner’s expertise, (ii) education or position in an hierarchy of an organization.In traditional scaffolding the social status of individuals is visible and easy to determine. For example, let us consider a lively face-to-face discussion between a novice in the construction field and a construction specialist, taking place in a construction company regarding possible solutions best suited to a construction problem. The construction specialist has a higher social status than the novice, due to a superior education, a broader experience and a higher position in the organizational hierarchy. Undoubtedly, while trying to reach a consensus on the best solution, the novice will be influenced by opinions of his tutor because of the latter’s convincing power.

In the online scaffolding context the best solution to a certain problem is usually found while interacting with many individuals and discussing different solutions provided by the online community. Usually, the online communities, in which informal learning takes place, are very large (people do not know each other) and they do not provide explicit indicators of users’ social status. It is difficult for users seeking for online help to relate the quality of the provided answers or solutions to the level of expertise or social status of individuals that proposed those solutions.

Indeed, there are Q&A sites, such as StackExchange, that reward their users with reputation scores based on their contributions. Based on the policies of such sites users get appropriate reputation scores for giving good answers, asking good questions or for voting on questions/answers of other users. It is obvious that high reputation users contribute high quality answers. But, there is a lack of research on the role of the high reputation users, i.e., if they also demonstrate high convincing power during the negotiation process, influencing opinions of other (low reputation) users. Thus, open research questions are: (i) if the presence of high reputation users, known as hubs, has an impact on spreading a single common opinion among other users and (ii) if relying too much on the reputation scores affects the quickness of the consensus building process?

What Layers did

Highlights

In the Layers project we investigated the influence of learners’ social status on their interactions with their peers and on negotiation processes in informal learning settings. We developed means to model the optimal extent of the influence of learners’ social status that speeds up the process of consensus building in a collaborative environment.

Description

Within the Layers project we investigated the influence of social status on negotiation processes that take place in collaboration networks by means of agent-based simulations. To that end, we simulated the diffusion of opinions in empirical collaboration networks, derived from six language editions (French, Spanish, Chinese, Japanese, German and English) of the Q&A site StackEchange, by taking into account both the network structure and the individual differences of people, reflected through their social status. In our study, the social status was represented by social reputation, which was awarded by the community.

We make use of the Naming Game model [3] that is well-known in the field of statistical physics and it has been successfully applied to model opinion diffusion and consensus building process. In the Naming Game model, agent-to-agent interactions take place based on predefined gaming rules. In particular, agents exchange their opinions and try to reach a consensus about the name of an unknown object. When all agents in the network agree on the name, the network is considered to have established a common opinion. According to the Naming Game rules, each agent has an inventory of words (names) that is initially empty. At each interaction step two random agents are chosen to communicate. One of them is declared as a speaker and the other one as a listener. The listener selects a word from her inventory and communicates it to the listener. If the listener’s inventory is empty, a new unique word is created and stored in the inventory. After communicating the word to the listener, two scenarios are possible:

  1. the word is not in the listener’s inventory: the word is added to listener’s inventory,

  2. otherwise, both speaker and listener agree on that word and remove all other words from their inventories: they agree on the selected word.

We extend the Naming Game model by incorporating a mechanism to configure the degree of the influence of social status on the network dynamics. We termed this mechanism the Probabilistic Meeting Rule [4]. Through parametrization, we are able to explore various scenarios from the opposite sides of the spectrum: (i) we can completely neglect the status by allowing any two individuals to exchange their opinions regardless of their social status (an egalitarian society), (ii) we can have opinions flowing only in one direction: from individuals with a higher social status to those with a lower social status (a stratified society), (iii) we can probabilistically model any situation in between these two extreme cases, i.e., a case in which opinions are very likely to flow from individuals with a higher social status to those with a lower social status but with small probability they can also flow into the other direction (a ranked society).

The situation after Layers

Highlights

By identifying social factors that lead to barriers and conflicts in collaborations and designing meaningful interventions (such as targeted user recommendations), Layers affected the trust building and negotiation processes in online scaffolding settings . Additionally, when the distribution of users’ social status in a collaboration network is known, Layers provides means to foster collaboration processes and meaning making in informal learning settings.

Description

The existing recommendation algorithms in Layers, which are mainly tag-based, are enhanced by providing more intelligent recommenders that additionally incorporate the social influence of the recommended users. We can identify users with high social status that correspond to users that are more capable of doing things during the online help-seeking process taking place in informal learning environments. Additionally, we can incorporate the influence of the social status on negotiation processes taking place in online scaffolding settings. Our work [5] shows that hubs (i.e., users with high degree and social status) are key to reaching consensus since they can distribute a single common opinion to a high number of other users.It is also evident that the influence of the social status of users involved in an online help-seeking process, to some extent, speeds up the consensus building while trying to agree on a best solution provided to a given problem.

We study the negotiation process in the whole network by determining the number of interactions needed for the whole network to achieve consensus and by investigating the direction of the communication flow. Our model facilitates the optimization of the influence of social statuses of persons involved in the collaboration, which in turn speeds up the process of consensus reaching.To illustrate the point, let’s consider the largest network that we investigated, the StackExchange English language network with 30,656 nodes and 192,983 edges, in which we performed 4 mil user interactions. Our experiments showed that when social status does not play a role, consensus is reached after 2.6 mil interactions, compared to the case in which social status plays a role, only 1.8 mil interactions are needed for all opinions in the network to converge to a common one. But, continuing to rely much more on the social status, consensus reaching starts to slow down. This indicates that there is a specific setting for each collaboration network in which social status speeds up the consensus building process.

We have analysed in detail the opinion flow between classes of users, defined based on their social status. So, we have high social status users, having a social status above the 90th percentile, and low status agents - below 90th percentile. We have found a setting in which the convergence of opinions (i.e., all users in the network agreed to one single opinion) is the fastest. Our findings showed that hubs (users that contribute the most of the content and possess high social status) are very important for the network to adopt a common opinion because they can distribute a single common opinion to a large number of other users.

But, the quickest consensus is achieved when (i) low status collaborators are allowed to freely exchange opinions between themselves (since this reduces the need for high status collaborators to interact with low status ones) and (ii) simultaneously there exists a communication barrier reducing the number of interactions of low status collaborators towards high status ones (since this reduces the variance in opinions of high status collaborators).

In our work [4] [5], this specific setting is achieved through parameterization. We measure the speed of the consensus building process based on the number of user interactions needed. In the specific setting mentioned above 20% less interactions were needed for the whole network to achieve consensus.

In terms of the constructor’s online forum [1], this would indicate that users that are trusted by others (based on peer evaluations and their status) should be able to always provide answers to others (users with low status). But, during the negotiation process (i.e., deciding which of the provided solutions is the best one), low status users should be prohibited to inflict their opinions on high status users. An example would be when low status users doubt a solution provided by a trusted user or argue that the provided solution is not worthy. This prolongs the negotiation process.

Learner
Organisation
Development

Impact that Layers created

Extended trust building and personal networks

Recommending learners based on their social status supports trust building among other learners. The social status of learners can be reflected through their social reputation they earned over their lifetime.

Within the Layers project we incorporated the influence of the social status on negotiation processes taking place in online scaffolding settings. This enables the identification of the more trusted answers (or opinions) about a problem solution.

In our study, we applied the Naming Game model (see section “What Layers did?”), which was already successfully used to model the opinion dynamics and consensus building in collaboration networks (see [4] and Consensus Building services).

Our findings showed that trusted users (known as hubs - they contribute the most of the content and possess high social status) are very important for the network to adopt a common opinion because they can distribute a single common opinion to a large number of other users.

Furthermore, our results reveal that there exists a specific setting, in which the optimal influence of the social status is evident, for each collaboration network. In this specific setting consensus building is reached quicker.

Improved the creation and use of learning resources

Learners having a high social status in a community are also characterized with collecting and producing high quality learning resources. So, driving them to collaborate and share their resources with other members of a learning community, improves the quality of learning resources in general.

Further Reading

Research papers: [4] [5] [1] [2]

Consensus building framework: Link

Slides presented at ASONAM’2015 in Paris, France:

The influence of social status on consensus building in collaboration networks by Ilire Hasani-Mavriqi

References

  1. K. Tammets, M. Laanpere, T. Ley, and K. Pata, “Identifying Problem-Based Scaffolding Patterns in an Online Forum for Construction Professionals,” in Scaling up Learning for Sustained Impact: 8th European Conference, on Technology Enhanced Learning, EC-TEL 2013, Paphos, Cyprus, September 17-21, 2013. Proceedings, D. Hernández-Leo, T. Ley, R. Klamma, and A. Harrer, Eds. Berlin, Heidelberg: Springer Berlin Heidelberg, 2013, pp. 526–531. DOI: 10.1007/978-3-642-40814-4_50
  2. K. Pata, P. Santos, and J. Burchert, “Social recognition provision patterns in professional Q&A forums in Healthcare and Construction,” Computers in Human Behavior, vol. 55, pp. 571–583, 2016.
  3. A. Baronchelli, M. Felici, V. Loreto, E. Caglioti, and L. Steels, “Sharp transition towards shared vocabularies in multi-agent systems,” Journal of Statistical Mechanics: Theory and Experiment, vol. 2006, no. 06, p. P06014, 2006.
  4. I. Hasani-Mavriqi, F. Geigl, S. C. Pujari, E. Lex, and D. Helic, “The Influence of Social Status on Consensus Building in Collaboration Networks,” in Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining 2015, 2015, pp. 162–169.
  5. I. Hasani-Mavriqi, F. Geigl, S. C. Pujari, E. Lex, and D. Helic, “The influence of social status and network structure on consensus building in collaboration networks,” Social Network Analysis and Mining, vol. 6, no. 1, p. 80, 2016.

Contributing Authors

Ilire Hasani-Mavriqi, Dominik Kowald, Tobias Ley, Elisabeth Lex