Successful online communities (e.g., Wikipedia, Yelp, and StackOverflow) can produce valuable content. However, many communities fail in their initial stages. Starting an online community is challenging because there is not enough content to attract a critical mass of active members. This paper examines methods for addressing this cold-start problem in datamining-bootstrappable communities by attracting non-members to contribute to the community. We make four contributions: 1) we characterize a set of communities that are “datamining-bootstrappable” and define the bootstrapping problem in terms of decision-theoretic optimization, 2) we estimate the model parameters in a case study involving the Open AI Resources website, 3) we demonstrate that non-members' predicted interest levels and request design are important features that can significantly affect the contribution rate, and 4) we ran a simulation experiment using data generated with the learned parameters and show that our decision-theoretic optimization algorithm can generate as much community utility when bootstrapping the community as our strongest baseline while issuing only 55% as many contribution requests.

The process of generating schematic maps of salient objects from a set of pictures of an indoor environment is challenging. It has been an active area of research as it is crucial to a wide range of context- and location-aware services, as well as for general scene understanding. Although many automated systems have been developed to solve the problem, most of them either require predefining labels or expensive equipment, such as RGBD sensors or lasers, to scan the environment. In this article, we introduce a prototype system to show how human computations can be utilized to generate schematic maps from a set of pictures, without making strong assumptions or demanding extra devices. The system requires humans (crowd workers from Amazon Mechanical Turks) to do simple spatial mapping tasks in various conditions, and their data are aggregated by filtering and clustering techniques that allow salient cues to be identified in the pictures and their spatial relations to be inferred and projected on a two-dimensional map. In particular, we tested and demonstrated the effectiveness of two methods that improved the quality of the generated schematic map: (1) We encouraged humans to adopt an allocentric representations of salient objects by guiding them to perform mental rotations of these objects and (2) we sensitized human perception by guided arrows superimposed on the imagery to improve the accuracy of depth and width estimation. We demonstrated the feasibility of our system by evaluating the results of schematic maps generated from indoor pictures taken from an office building. By calculating Riemannian shape distances between the generated maps to the ground truth, we found that the generated schematic maps captured the spatial relations well. Our results showed that the combination of human computations and machine clustering could lead to more-accurate schematized maps from imagery. We also discuss how our approach may have important insights on methods that leverage human computations in other areas.

E-petitioning has become one of the most important and popular forms of online activism. Although e-petition success is driven by user behavior, users have received relatively little study by HCI and social computing researchers. Drawing from theoretical and empirical work in analogous social computing systems, we identify two potentially competing theories about the trajectories of users in e-petition platforms: (1) "power" users in social computing systems are born, not made; and (2) users mature into "power" users. In a quantitative analysis of data from Change.org, one of the largest online e-petition platforms, we test and find support for both theories. A follow-up qualitative analysis shows that not only do users learn from their experience, systems also "learn" from users to make better recommendations. In this sense, we find that although power users are "born," they are also "made" through both processes of personal growth and improved support from the system.

Hundreds of thousands of crowdfunding campaigns have been launched, but more than half of them have failed. To better understand the factors affecting campaign outcomes, this paper targets the content and usage patterns of project updates -- communications intended to keep potential funders aware of a campaign's progress. We analyzed the content and usage patterns of a large corpus of project updates on Kickstarter, one of the largest crowdfunding platforms. Using semantic analysis techniques, we derived a taxonomy of the types of project updates created during campaigns, and found discrepancies between the design intent of a project update and the various uses in practice (e.g. social promotion). The analysis also showed that specific uses of updates had stronger associations with campaign success than the project's description. Design implications were formulated from the results to help designers better support various uses of updates in crowdfunding campaigns.

Feedback on designs is critical for helping users iterate toward effective solutions. This paper presents Voyant, a novel system giving users access to a non-expert crowd to receive perception-oriented feedback on their designs from a selected audience. Based on a formative study, the system generates the elements seen in a design, the order in which elements are noticed, impressions formed when the design is first viewed, and interpretation of the design relative to guidelines in the domain and the user's stated goals. An evaluation of the system was conducted with users and their designs. Users reported the feedback about impressions and interpretation of their goals was most helpful, though the other feedback types were also valued. Users found the coordinated views in Voyant useful for analyzing relations between the crowd's perception of a design and the visual elements within it. The cost of generating the feedback was considered a reasonable tradeoff for not having to organize critiques or interrupt peers.

LinkedIn is the world's largest professional network, with over 300 million members. One of the primary activities on the site is people search, for which LinkedIn members are both the users and the corpus. This paper presents insights about people search behavior on LinkedIn, based on a log analysis and a user study. In particular, it examines the role that network distance plays in name searches and non-name searches. For name searches, users primarily click on only one of the results, and closer network distance leads to higher click-through rates. In contrast, for non-name searches, users are more likely to click on multiple results that are not in their existing connections, but with whom they have shared connections. The results show that, while network distance contributes significantly to LinkedIn search engagement in general, its role varies dramatically depending on the type of search query.

Visually impaired people have difficulty in walking freely because of the obstacles or the stairways along their walking paths, which can lead to accidental falls. Many researchers have devoted to promoting safe walking for visually impaired people by using smartphones and computer vision. In this research we propose an alternative approach to achieve the same goal - we take advantage of the power of crowdsourcing with machine learning. Specifically, by using smartphones carried by a vast amount of visually normal people, we can collect the tri-axial accelerometer data along with the corresponding GPS coordinates in large geographic areas. Then, machine learning techniques are used to analyze the data, turning them into a special topographic map in which the regions of outdoor stairways are marked. With the map installed in the smartphones carried by the visually impaired people, the Android App we developed can monitor their current outdoor locations and then enable an acoustic alert whey they are getting close to the stairways.

This paper studied how social transparency and different peer-dependent reward schemes (i.e., individual, teamwork, and competition) affect the outcomes of crowdsourcing. The results showed that when social transparency was increased by asking otherwise anonymous workers to share their demographic information (e.g., name, nationality) to the paired worker, they performed significantly better. A more detailed analysis showed that in a teamwork reward scheme, in which the reward of the paired workers depended only on the collective outcomes, increasing social transparency could offset effects of social loafing by making them more accountable to their teammates. In a competition reward scheme, in which workers competed against each other and the reward depended on how much they outperformed their opponent, increasing social transparency could augment effects of social facilitation by providing more incentives for them to outperform their opponent. The results suggested that a careful combination of methods that increase social transparency and different reward schemes can significantly improve crowdsourcing outcomes.

Peer consistency evaluation is often used in games with a purpose (GWAP) to evaluate workers using outputs of other workers without using gold standard answers. Despite its popularity, the reliability of peer consistency evaluation has never been systematically tested to show how it can be used as a general evaluation method in human computation systems. We present experimental results that show that human computation systems using peer consistency evaluation can lead to outcomes that are even better than those that evaluate workers using gold standard answers. We also show that even without evaluation, simply telling the workers that their answers will be used as future evaluation standards can significantly enhance the workers' performance. Results have important implication for methods that improve the reliability of human computation systems.

We created a spatial location identification task (SpLIT) in which workers recruited from Amazon Mechanical Turk were presented with a camera view of a location, and were asked to identify the location on a two-dimensional map. In cases where these cues were ambiguous or did not provide enough information to pinpoint the exact location, workers had to make a best guess. We tested the effects of two reward schemes. In the “ground truth” scheme, workers were rewarded if their answers were close enough to the correct locations. In the “majority vote” scheme, workers were told that they would be rewarded if their answers were similar to the majority of other workers. Results showed that the majority vote reward scheme led to consistently more accurate answers. Cluster analysis further showed that the majority vote reward scheme led to answers with higher reliability (a higher percentage of answers in the correct clusters) and precision (a smaller average distance to the cluster centers). Possible reasons for why the majority voting reward scheme was better were discussed.

Crowdsourcing and machine learning are both useful techniques for solving difficult problems (e.g., computer vision and natural language processing). In this paper, we propose a novel method that harnesses and combines the strength of these two techniques to better analyze the features and the sentiments toward them in user reviews. To strike a good balance between reducing information overload and providing the original context expressed by review writers, the proposed system (1) allows users to interactively rank the entities based on feature-rating, (2) automatically highlights sentences that are related to relevant features, and (3) utilizes implicit crowdsourcing by encouraging users to provide correct labels of their own reviews to improve the feature-sentiment classifier. The proposed system not only helps users to save time and effort to digest the often massive amount of user reviews, but also provides real-time suggestions on relevant features and ratings as users generate their own reviews. Results from a simulation experiment show that leveraging on the crowd can significantly improve the feature-sentiment analysis of user reviews. Furthermore, results from a user study show that the proposed interface was preferred by more participants than interfaces that use traditional noun-adjective pair summarization, as the current interface allows users to view feature-related information in the original context.