Multiplatform Reactive Crowdsourcing based on Social Networks – WWW2013

In this post I want to report on our paper on Reactive Crowdsourcing presented at the WWW 2013 conference in Rio de Janeiro, Brasil.

Here is a quick summary of motivation and idea, together with some relevant materials:

Need for control

We believe that an essential aspect for building effective crowdsourcing computations is the ability of “controlling the crowd”, i.e. of dynamically adapting the behaviour of the crowdsourcing systems as response to the quantity and quality of completed tasks or to the availability and reliability of performers.
This new paper focuses on a machinery and methodology for deploying configurable, cross-platform, and adaptive crowdsourcing campaigns through a model-driven approach.

Control through declarative active rules

In the paper we present an approach to crowdsourcing which provides powerful and flexible crowd controls. We model each crowdsourcing application as composition of elementary task types and we progressively transform these high level specifications into the features of a reactive execution environment that supports task planning, assignment and completion as well as performer monitoring and exclusion. Controls are specified as declarative, active rules on top of data structures which are derived from the model of the application; rules can be added, dropped or modified, thus  guaranteeing maximal exibility with limited effort. The paper applies modeling practices (as also explained in our book on model-driven software engineering).

Here is the presentation thatAlessandro Bozzon gave at WWW 2013:

Reactive crowdsourcing presentation on slideshare.

Prototype and experiments

We have a prototype platform that implements the proposed framework.  We have done extensive experiments with it. Our experimentations with different rule sets demonstrate how simple changes to the rules can substantially affect time, effort and quality involved in crowdsourcing activities.

Here is a short video demonstrating our approach through the current prototype (mainly centered on the crowdsourcing campaign configuration phase):

Paper and related activities

The paper can be downloaded for free from the ACM Digital Library through this link:
http://dl.acm.org/citation.cfm?id=2488403 (or alternatively from the WWW site)

The paper is a follow-up of our WWW2012 paper on Crowdsearcher, which focused on exploiting social networks and crowdsourcing platforms for improving search.
The paper nicely combines with another recent contribution of ours, presented at EDBT 2013, on finding the right crowd of experts on social networks for addressing a specific problem.

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

My invited post on Modeling Social Web Apps (on www.modeling-languages.com)

It’s with great pleasure that I announce my invited post on the modeling-languages.com blog, curated by Jordi Cabot.
First, I’m glad he invited me. Second, I’m happy that he asked for a post on Social-enabled Web application modeling.
I mean, we all see how social technologies are transforming our life. And yet, the modeling community and the software engineering community at large are paying very limited attention to this phenomenon.
That’s why I decided to address the problem by proposing a model-driven approach that is specifically focused on the development of Web applications that exploit social features, and my invited post is exactly focusing on this.
Basically, the proposal is to go from requirement specification down to static and dynamic design and to code generation of social applications with a pattern-based approach, which exploits goal-based requirement specification, UML modeling, and WebML models (enriched with social-specific primitives) for describing the user interaction. This image intuitively summarizes the idea:

You can find more details in the invited post on Social web application modeling on Jordi’s blog, and also check out the video that summarizes the approach:

 To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

Efficient Subgraph Matching – Keynote by V.S. Subrahmanian at ASONAM 2012

V.S. Subrahmanian (Univ. of Maryland)
As part of the Data Management in the Social Semantic Web workshop (DMSSW workshop) at the ASONAM 2012 conference in Istanbul, Turkey, V.S. Subrahmanian (University of Maryland) gave an interesting talk on efficient subgraph matching on (social) networks.
Queries are defined as graphs themselves, with some of the places defined as constants and some defined as variables.
The complexity of queries over graphs is high, due to the large number of joins to be performed even in case of fairly simple queries.
The size of the query is typically relatively small with respect to the entire dataset (network). The proposed approaches are useful for scale of at least tens of millions of nodes in the network.

How to work on disk

The mechanism implemented is called DOGMA index and applies an algorithm called K-merge.
The algorithm builds a hierarchical index where I put at most K nodes of the graph in each index item. For obtaining that, I merge together connected nodes.You can do that randomly or more intelligently by trying to minimizing connections between nodes in different index items.
Example of DOGMA Index, where nodes of the original network (at the bottom) are merged in higher level representations in the level above (in this example, K = 4, since we have 4 nodes in each index position).
I don’t want to build the index by partitioning the whole graph, because it’s painful for large graphs. 
I start from a G0 graph, and I merge nodes until I get G1, G2, Gn graphs, each of them is more or less half the size of the previous, until Gn has K nodes or less. Then, I build the dogma index over Gn.
For the query, I can use a basic approach: identify the variable nodes that are immediately close to a constant node, and then find the possible satisfying values for those variables, starting from the constants. I can apply conditions considering distance constraints between constants and variables, as well as between candidate variable names. To allow this, I also save in every node of the index the distance of closest node in the other nodes. 

How to work on the cloud

This approach has been also implemented in the cloud, through the so called COSI Architecture, assuming a cloud of k+1 computing nodes. The implementation of the edge cuts that generates the index must be very quick and produce fairly good cuts (but not necessarily optimal).
The image below lists some of the references to S.V. works on the topic.
Some references to V.S. Subrahmanian works on subgraph matching.

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

Answering search queries with CrowdSearcher – WWW2012

Our paper:

 Answering search queries with CrowdSearcher
has been accepted and presented at WWW 2012 in Lyon.

Here is the abstract:
Web users are increasingly relying on social interaction to complete and validate the results of their search activities. While search systems are superior machines to get world-wide information, the opinions collected within friends and expert/local communities can ultimately determine our decisions: human curiosity and creativity is often capable of going much beyond the capabilities of search systems in scouting “interesting” results, or suggesting new, unexpected search directions. Such personalized interaction occurs in most times aside of the search systems and processes, possibly instrumented and mediated by a social network; when such interaction is completed and users resort to the use of search systems, they do it through new queries, loosely related to the previous search or to the social interaction. In this paper we propose CrowdSearcher, a novel search paradigm that embodies crowds as first-class sources for the information seeking process. CrowdSearcher aims at filling the gap between generalized search systems, which operate upon world-wide information – including facts and recommendations as crawled and indexed by computerized systems – with social systems, capable of interacting with real people, in real time, to capture their opinions, suggestions, emotions. The technical contribution of this paper is the discussion of a model and architecture for integrating computerized search with human interaction, by showing how search systems can drive and encapsulate social systems. In particular we show how social platforms, such as Facebook, LinkedIn and Twitter, can be used for crowdsourcing search-related tasks; we demonstrate our approach with several prototypes and we report on our experiment upon real user communities.

The full paper is available here:
http://dl.acm.org/citation.cfm?id=2187836.2187971&coll=DL&dl=ACM

The presentation I gave is this one:

The demo video can be found on YouTube:


To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

Social Informatics workshop: Social networking at work with human factors and technical platforms in mind

Today I’m attending an extremely interesting workshop on Social Informatics at University of Trento. The workshop gathers a diverse audience spanning from computer science to cognitive science and health.
The program of the day included the following presentations:

Supporting Social Interactions for Older Adults. Cristhian Parra, University of Trento, Italy
This talk focused on ways for encouraging social sharing and interactions between elders, and on which could be the fields where elders could provide and get higher benefits in online socialization.

Active Lifestyle applications and motivation instruments. Patricia Silveira, University of Trento, Italy
This talk discussed the role of serious games in preserving elders physical and mental functions and health.

Towards a crowdsourcing platform for elders. Pavel Kucherbaev, University of Trento, Italy. Neocogita S.r.l. – UNITN Spin-off in Cognitive Training. Nicola De Pisapia, University of Trento, Italy
This talk focused on understanding which are the cognitive capabilities that need to be continuously trained and maintained, also through games.
 Experience sharing: LiquidGalleries and ComeAlong. Beatrice Valeri, University of Trento, Italy
This talk presented two interesting experiences of sharing: LiquidGalleries is a flexible and social mobile app that allows a personalized and delocalized experience when visiting art exhibitions and museums. People can tag preferred art pieces in advance, get content on the fly, or tag pieces while visiting and sharing them or reading about them later at home.

Knowledge Spaces: Supporting Knowledge and Experience Sharing. Marcos Baez, University of Trento, Italy
This talk presented a set of methodological guidelines for defining experience sharing applications and showed them at work within the scenario of scientific publication sharing.

Sensing social interactions through smart phones. Aleksandar Matic, University of Trento and Create-net, Italy
This talk presented some techniques for detecting social activities in real world considering physical proximity, based on smart phone technologies (as opposed to dedicated hw used in other projects).

Civic Media Platforms. Maurizio Teli and Stefano De Paoli, ahref Foundation (Trento), Italy
The talk focused on platforms that aim at increasing the citizen participation to a common good target in local or government communities or activities, including also citizen journalism. The addressed problem is how to build a social community in this scenario, considering also possible biases (political or social) added. An example is the Timu platfom, focusing on storytelling.

BPM4Crowd.  Stefano Tranquilini, University of Trento, ItalyThis presentation suggests some basic technical support to the development of social applications. The idea is to provide a high-level abstraction access to the developer, so as to ease the development of applications. This is demonstrated with an online dating application implemented within a social network.
 
Convergence of social networking, search and business processes. Marco Brambilla, Politecnico di Milano, Italy
This was my talk and included an overview on Social BPM and CrowdSearch. I will post some slides very soon.

 Recommending content for basic and high education. Rosa Alarcon, Pontificia Universidad Catolica de Chile, Chile
This talk was about understanding the information overload in the educational market.
Existing content repositories for elearning are not sufficient. The standard ways for describing them (e.g., SCORM) are quite poor too.
They try to apply recommendation techniques for teachers and they end up with some initial results, including the fact that surprisingly teachers should not be clustered based on their specialization.

Interactive Experiences for supporting elderly or impaired young people. Paolo Massa, FBK, I3, Italy
This talk was about a few experiences on  applying technological tools for facilitating tasks and socialization of autistic boys and people with mental disorders.
Other activities are about biases in translation and editing on the web, especially on public sites like wikipedia (e.g., see http://manypedia.com or the wikitrip project), considering gender and political problems.

Credibility evaluation of Web content. Adam Wierzbicki, Polish-Japanese Institute of Information Technology, Poland
This speech presented an interesting corpus of Web credibility assessment. The assessment is done by asking people to assess credibility and cleanness of pages.
Game theory has been used also for classifying credibility. This research is a joint work with some Swiss institutes, including EPFL.
The talk also addressed the roles of wikipedia. It’s not just an encyclopedia, its a knowledge community, a collaborative innovation network, a community of prosumers, a model for the knowledge economy. But information about it is limited and social behaviours are critical (the environment is getting more and more unpleasant and closed). Overall, it cannot be defined a social centric platform.
Finally, the speech focused on serious social games for elders to motivate social interaction. A good paradigm is gaming based on gossiping because it grants trust, social norms, and ties.

Other speeches I could not attend included:

  • Cheating in online games / trust and reputation. Stefano De Paoli, ahref Foundation (Trento), Italy
  • Reseval Mash. Muhammad Imran, University of Trento, Italy
  • Social Processes over Social Networks. Juan Jose Jara, University of Trento, Italy
  • Discovery of composition knowledge for mashup development, process mining and BPM (BI). Carlos Rodriguez, University of Trento, Italy
  • Social Search and recommendation engine for scientific publications. Daniil Mirilenka, University of Trento, Italy
  • Overview of DALi project. Christopher Raphael Wilkinson, University of Trento, Italy

This was a great and multi-disciplinary experience, very well in line with our research project Search Computing, Cubrik, and BPM4People.

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).