Task recommendation in reward-based crowdsourcing systems

Kurup, Ayswarya R; Sajeev, G P

doi:10.1109/icacci.2017.8126055

Cited by 23 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another related area is online crowdsourcing, where the aim is to allocate tasks published by a requester to an online crowd. [18,33] employ a probabilistic matrix factorization (PMF) to recommend suitable tasks to crowdworkers based on their previous activities, performance, and preferences. They handle the cold start problem by utilizing predefined categories (e.g., sentiment analysis, translation, image labeling) as additional features to improve the recommendation accuracy.…”

Section: Recommending Tasks To Communities and Crowds Onlinementioning

confidence: 99%

Learning to Recommend Items to Wikidata Editors

Alghamdi¹,

Shi²,

Simperl³

2021

Preprint

View full text Add to dashboard Cite

Wikidata is an open knowledge graph built by a global community of volunteers. As it advances in scale, it faces substantial challenges around editor engagement. These challenges are in terms of both attracting new editors to keep up with the sheer amount of work, and retaining existing editors. Experience from other online communities and peer-production systems, including Wikipedia, suggests that personalised recommendations could help, especially newcomers, who are sometimes unsure about how to contribute best to an ongoing effort. For this reason, we propose a recommender system WikidataRec for Wikidata items. The system uses a hybrid of content-based and collaborative filtering techniques to rank items for editors relying on both item features and item-editor previous interaction. A neural network, named neural mixture of representations, is designed to learn fine weights for the combination of item-based representations and optimize them with editor-based representation by item-editor interaction. To facilitate further research in this space, we also create two benchmark datasets, a general-purpose one with 220, 000 editors responsible for 14 million interactions with 4 million items, and a second one focusing on the contributions of more than 8, 000 more active editors. We perform an offline evaluation of the system on both datasets with promising results. Our code and datasets are available at https://github.com/WikidataRec-developer/Wikidata_Recommender.

show abstract

Section: Recommending Tasks To Communities and Crowds Onlinementioning

confidence: 99%

Learning to Recommend Items to Wikidata Editors

Alghamdi¹,

Shi²,

Simperl³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The relation REVIEW is created by duplicating each tuple in VEHICLE 10 times, then adding a sentiment attribute to generate random values. 34 (ii) The 2014 car models with real dataset Auto2 are used. The real crowd is used to evaluate the images, reviews, and car specifications of the dataset.…”

Section: Dataset Representationmentioning

confidence: 99%

An efficient privacy‐preserving model based on OMFTSA for query optimization in crowdsourcing

Renukadevi¹,

Anita

Geetha³

2021

Concurrency and Computation

View full text Add to dashboard Cite

Crowdsourcing is now one of the most important and transformative paradigms, with great success in a variety of application tasks. Crowdsourcing obtains knowledge and information to solve cognitive or intelligence-intensive tasks from an evolving group of participants via the Internet. Unfortunately, providing a hard privacy guarantee and query optimization is incompatible when a higher task acceptance rate needs to be accomplished and this case is common in most existing crowdsourcing solutions. The state of art systems suffered from different complexities such as lack of crowdsourcing optimization techniques, increased cost, latency, security, and scalability issues. In this paper, we have proposed a crowdsourcing model to optimize the cost and latency, issues that occur while query optimization using the Moth Flame and Tunicate Swarm Algorithm (MF-TSA). The TSA algorithm is added to the MF algorithm to enhance its exploitation capability and yield fast convergence. The data privacy concerns of the worker and the requestor are addressed using homomorphic encryption that simultaneously enhances the efficiency of the crowdsourcing framework. The main aim of this work is to optimize the cost and latency for query plan selection along with security. Initially, the homomorphic encryption model is used to encrypt the data. In query design, two kinds of crowd-controlled administrators, that is, Crowd Powered Selection (CSelect) and Crowd Powered Join (CJoin) are connected for assessing query. The proposed framework utilizes MF-TSA to optimize the selection and join queries with low cost and latency. Finally, the experimental results demonstrate better query optimization performance than other existing algorithms such as sequential, parallel, andcrowdsourcing, moth flame-based tunicate swarm algorithm, query, select and join queries INTRODUCTIONRecently, large-scale user representations of extensive collaboration have been made available through the Internet via various types of technologies and mechanisms. The collection of text documents and data source connections is an important part of the World Wide Web (WWW). The emerging crowdsourcing platform has been authorized to collect and distribute human computing for a wide range of tasks. 1 To generate an accurate and efficient solution, crowdsourcing targets a massive audience. The consensus is shared between the crowdsourcing system and implementations, such as employees who participate and delegate tasks that differ. When it comes to tasks such as answering questions, some workers are smarter and faster than others. 2,3

show abstract

“…Mayank [11] presented a hybrid algorithm for optimal resource allocation under constrained task schedule. Systems with crowd sourcing act as distributed problem solving platforms, where task channeling into crowd takes place these solutions are discussed by Kurup [12].…”

Section: Comparison With Related Workmentioning

confidence: 99%

An Optimal Algorithm to Maximize Vaccine Production in a Distributed Environment

Saran

Vignesh

Sundar

et al. 2021

2021 12th International Conference on Computing Communication and Networking Technologies (ICCCNT)

View full text Add to dashboard Cite

Pandemics have a devastating effect on lives across the globe as evinced by the ongoing COVID-19. Production of vaccine, medicines and other critical supplies such as oxygen are indispensable and are to be expedited. This work examines the problem of maximizing the manufacturing of critical supplies within a target stipulated time by various manufacturing units. Two scenarios are considered in vaccine production: centralized and distributed. In the former case, the manufacturing units are closely located and transport the raw materials with no delay. In the latter case, the units are located geographically apart and transport of raw materials takes a finite time causing delays in vaccine production. The main contribution of the paper lies in the elucidation of optimal algorithms for both scenarios that take the dependencies between the ingredients into consideration. The dependencies have a constraining effect on the order of manufacturing of both the raw materials and the end products referred to as targets and hence careful consideration is necessary to maximize production. The optimality is measured by an incentive-based scheme that seeks to reward the faster development of targets.

show abstract

Task recommendation in reward-based crowdsourcing systems

Cited by 23 publications

References 21 publications

Learning to Recommend Items to Wikidata Editors

Learning to Recommend Items to Wikidata Editors

An efficient privacy‐preserving model based on OMFTSA for query optimization in crowdsourcing

An Optimal Algorithm to Maximize Vaccine Production in a Distributed Environment

Contact Info

Product

Resources

About