检索结果-内蒙古大学图书馆

Real-time recommendation algorithms for crowdsourcing systems

Applied Computing and Informatics 2017年第1期13卷 47-56页

作者： Safran, Mejdl Che, Dunren Computer Science Department Southern Illinois University Carbondale IL United States Computer Science Department King Saud University Riyadh Saudi Arabia

Crowdsourcing has become a promising paradigm for solving tasks that are beyond the capabilities of machines alone via outsourcing tasks to online crowds of people. Both requesters and workers in crowdsourcing systems confront a flood of data coming along with the vast amount of tasks. Fast, on-the-fly recommendation of tasks to workers and workers to requesters is becoming critical for crowdsourcing systems. Traditional recommendation algorithms such as collaborative filtering no longer work satisfactorily because of the unprecedented data flow and the on-the-fly nature of the tasks in crowdsourcing systems. A pressing need for real-time recommendations has emerged in crowdsourcing systems: on the one hand, workers want effective recommendation of the top-k most suitable tasks with regard to their skills and preferences, and on the other hand, requesters want reliable recommendation of the top-k best workers for their tasks in terms of workers’ qualifications and accountability. In this article, we propose two real-time recommendation algorithms for crowdsourcing systems: (1) TOP-K-T that computes the top-k most suitable tasks for a given worker and (2) TOP-K-W that computes the top-k best workers to a requester with regard to a given task. Experimental study has shown the efficacy of both algorithms. © 2016 The Authors

关键词： Crowdsourcing recommendation algorithms Task recommendation Top-k tasks Top-k workers Worker recommendation

来源：评论

学校读者我要写书评

暂无评论

Incorporating Consumer Browse Data: Extended Item-Based Top-K recommendation algorithms

Incorporating Consumer Browse Data: Extended Item-Based Top-...

引用

International Conference on Information Technology and Applications (ITA)

作者： Du, Xiaomeng Su, Meng Peking Univ Guanghua Sch Management Beijing 100871 Peoples R China

ISBN: (纸本)9781479928767

Personalized recommendation systems are becoming increasingly popular in e-commerce. One of the core recommendation algorithms is item-based recommendation algorithm, which recommend by finding products that are similar to other products using consumers' purchase data. However, purchase data contains too little information to reveal the real preference of consumers. The purpose of this paper is to extend the item-based top-k recommendation algorithm by incorporating consumers' browse data, which contains more useful information in revealing the preference of consumers. First, we extend the cosine-based method of modeling the similarity matrix between two products and construct an extended model subjected to varying conditions in terms of browse and purchase combination. Then we test our extended algorithm on a simulated data set and apply it to a real world data set provided by an online B2C company. Both the results of the simulation data and the real data show that our extended algorithm performs significantly better than the traditional item-based algorithm, in terms of hit rate (HR) and average reciprocal hit rank (ARHR). Specifically, our extended algorithm generates 113,089 (4349.58%) more sales than random algorithms and 25,335 (28.86%) more sales than traditional cosine-based algorithms. We discuss the theoretical and managerial implications in the conclusion part.

关键词： recommendation algorithms personalized recommendation systesms consumer browse data item-based top-k recommendation algorithms e-commerce

来源：评论

学校读者我要写书评

暂无评论

Context2Rec: Leveraging comment text semantics and sequential features for enhanced recommendation systems

引用

PATTERN RECOGNITION LETTERS 2025年 189卷 136-142页

作者： Luo, Wenlong Zhang, Li Cui, Yachao Jin, Yanzheng Shaanxi Normal Univ Sch Comp Sci Xian 710000 Shaanxi Peoples R China Qinghai Univ Sch Comp Technol & Applicat Xining 810016 Qinghai Peoples R China Natl Univ Singapore Sch Comp Singapore 119077 Singapore

The integration of comment text into recommendation methods aims to leverage user-generated content as an auxiliary data source to enhance recommendation quality. A deep learning model, Context2Rec, is proposed to improve product interactivity and address the multifaceted needs of user feature learning. The model utilizes user comments to extract crucial word order and contextual relationships within the text, thereby enhancing semantic representation through pre-training. By innovatively combining comment text features with sequential data, Context2Rec significantly enhances the performance of recommendation systems. The model captures dynamic shifts in user preferences by employing BERT for semantic understanding and incorporating item2vec for association-based feature learning. Experimental evaluations across multiple datasets demonstrate that Context2Rec outperforms existing algorithms, providing an effective approach for personalized recommendation services. The experimental results indicate that Context2Rec achieves an average reduction in MSE of 14.3% on the Amazon dataset, 16.2% on the Yelp dataset, and 12.6% on the Beer dataset, while showing superior recommendation performance, particularly in sparse data scenarios.

关键词： recommendation algorithms BERT Attention mechanisms Review text

来源：评论

学校读者我要写书评

暂无评论

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS 2025年第3期16卷 334-342页

作者： Lu, Wei Cai, Xiaodong Li, Minghui Guilin Univ Elect Technol Sch Informat & Commun Guilin Peoples R China

In recommendation systems, negative sampling strategies are crucial for the calculation of contrastive learning loss. Traditional random negative sampling methods may lead to insufficient quality of negative samples during training, thereby affecting the convergence and performance of the model. In addition, the Bayesian Personalized Ranking (BPR) loss function usually converges slowly and is prone to falling into suboptimal local solutions. To address the above problems, this paper proposes a recommendation algorithm based on popularity-corrected sampling and improved contrastive loss. First, a dynamic negative sampling method with popularity correction is proposed, which reduces the impact of item popularity distribution bias on model training and dynamically screens out negative samples to improve the quality of model recommendations. Second, an improved contrastive loss is proposed, which selects the most challenging negative samples and introduces a boundary threshold to control the sensitivity of the loss, enabling the model to focus more on samples that are difficult to distinguish and further optimize the recommendation effect. Experimental results on the Amazon-Book, Yelp2018, and Gowalla datasets show that the proposed model significantly outperforms mainstream state-of-the-art models in recommendation tasks. Specifically, the Recall metric, which reflects model accuracy, improves by 16.8%, 12.9%, and 5.72% respectively on these three datasets. The NDCG metric, which measures ranking quality, increases by 20.7%, 16.4%, and 7.76% respectively. These results confirm the effectiveness and superiority of the recommendation algorithm across different scenarios. Compared with baseline models, it demonstrates stronger adaptability in complex situations, such as the sparse dataset Gowalla and the long-tail distribution dataset Amazon-Book, with the highest improvement in core metrics exceeding 20%.

关键词： recommendation algorithms contrast loss difficult negative samples object popularity bias

来源：评论

学校读者我要写书评

暂无评论

Qualitative analysis of user-based and item-based prediction algorithms for recommendation agents

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2005年第7期18卷 781-789页

作者： Papagelis, M Plexousakis, D Fdn Res & Technol Hellas Inst Comp Sci GR-71110 Iraklion Greece Univ Crete Dept Comp Sci GR-71409 Iraklion Greece

recommendation agents employ prediction algorithms to provide users with items that match their interests. In this paper, several prediction algorithms are described and evaluated, some of which are novel in that they combine user-based and item-based similarity measures derived from either explicit or implicit ratings. Both statistical and decision-support accuracy metrics of the algorithms are compared against different levels of data sparsity and different operational thresholds. The first metric evaluates the accuracy in terms of average absolute deviation, while the second evaluates how effectively predictions help users to select high-quality items. The experimental results indicate better performance of item-based predictions derived from explicit ratings in relation to both metrics. Category-boosted predictions lead to slightly better predictions when combined with explicit ratings, while implicit ratings, in the context that have been defined in this paper, perform much worse than explicit ratings. (c) 2005 Elsevier Ltd. All rights reserved.

关键词： recommendation algorithms collaborative filtering similarity measures

来源：评论

学校读者我要写书评

暂无评论

Constructing recommendation Systems for Effective Health Messages Using Content, Collaborative, and Hybrid algorithms

引用

ANNALS OF THE AMERICAN ACADEMY OF POLITICAL AND SOCIAL SCIENCE 2015年第1期659卷 290-306页

作者： Cappella, Joseph N. Yang, Sijia Lee, Sungkyoung Univ Penn Annenberg Sch Commun Commun Philadelphia PA 19104 USA Univ Penn Annenberg Sch Commun Philadelphia PA 19104 USA Univ Missouri Sch Journalism Columbia MO 65211 USA

Theoretical and empirical approaches to the design of effective messages to increase healthy and reduce risky behavior have shown only incremental progress. This article explores approaches to the development of a recommendation system for archives of public health messages. recommendation systems are algorithms operating on dense data involving both individual preferences and objective message features. Their goal is to predict ratings for items (i.e., messages) not previously seen by the user on content similarity, prior preference patterns, or their combination. Standard approaches to message testing and research, while making progress, suffer from very slow accumulation of knowledge. This article seeks to leapfrog conventional models of message research, taking advantage of modeling developments in recommendation systems from the commercial arena. After sketching key components in developing recommendation algorithms, this article concludes with reflections on the implications of these approaches in both theory development and application.

关键词： recommendation algorithms content-based approaches collaborative filtering hybrid approaches health message design

来源：评论

学校读者我要写书评

暂无评论

Design and evaluation of an English speaking recommender system using word networks and context-aware techniques

引用

ENTERTAINMENT COMPUTING 2025年 52卷

作者： Ding, Zhenyue East China Normal Univ Dept Educ Shanghai 200062 Peoples R China

Speaking is an important part of language communication, but mastering speaking skills and rich vocabulary is a challenge for non-native speakers. To address this problem, the study designs an English speaking recommendation system that incorporates word networks as well as user-based context-aware and semantic-aware techniques. The word network identifies associations between words by analyzing extensive English-speaking learning resources. While the user-based context-aware and semantic-aware techniques generate recommendation results by analyzing the similarity of users. Meanwhile, Precision, Recall, and F1 were selected for comparison. Precision indicates the fraction of cases labeled as positive that are truly positive. Recall indicates the fraction of actual positive cases that are correctly predicted as positive. The F1 score is the average of precision and recall, calculated as their harmonic mean. The results show that each recommendation evaluation index of the spoken English recommendation system shows a tendency of increasing and then decreasing with the increase of the interest similarity weight value Wic. Meanwhile, the individual recommendation evaluation indexes of the English speaking recommender system decrease with the increase of alpha value. Among them, when the alpha value is 0.5, the system has the best recommendation results with Precision, Recall, and F1 as high as 75.69 %, 25.81 %, and 37.48 %, respectively. In addition, each interface of the system provides good response time under high concurrency. In particular, the response time of the login interface is as low as 120 ms and the response time of the recommendation interface is as high as 328 ms on average. Compared to current recommendation systems, the average response time has been improved by at least 50 ms. It shows that the English speaking recommendation system designed by the Institute has good performance advantages and feasibility. It provides reliable technical support for mod

关键词： Word networks recommendation algorithms Spoken English recommendation system design User similarity

来源：评论

学校读者我要写书评

暂无评论

An approach to sociotechnical transparency of social media algorithms using agent-based modelling

引用

AI and ethics 2025年第2期5卷 1827-1845页

作者： Anna Gausen Ce Guo Wayne Luk Imperial College London London UK.

The recommendation algorithms on social media platforms are hugely impactful, they shape information flow and human connection on an unprecedented scale. Despite growing criticism of the social impact of these algorithms, they are still opaque and transparency is an ongoing challenge. This paper has three contributions: (1) We introduce the concept of . This can be defined as transparency approaches that consider both the technical system, and how it interacts with users and the environment in which it is deployed. We propose sociotechnical approaches will improve the understanding of social media algorithms for policy-makers and the public. (2) We present an approach to sociotechnical transparency using agent-based modelling, which overcomes a number of challenges with existing approaches. This is a novel application of agent-based modelling to provide transparency into how the recommendation algorithm prioritises different curation signals for a topic. (3) This agent-based model has a novel implementation of a multi-objective recommendation algorithm that is calibrated and empirically validated with data collected from X, previously Twitter. We show that agent-based modelling can provide useful insights into how the recommendation algorithm prioritises different curation signals. We can begin to explore whether the priorities of the recommendation algorithm align with what platforms say it is doing and whether they align with what the public want.

关键词： Agent-based modelling recommendation algorithms Social media Transparency

来源：评论

学校读者我要写书评

暂无评论

Tuning metadata for better movie content-based recommendation systems

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2015年第17期74卷 7015-7036页

作者： Soares, Marcio Viana, Paula INESC TEC P-4200465 Oporto Portugal Polytech Inst Porto ISEP IPP Sch Engn P-4200072 Oporto Portugal

The increasing number of television channels, on-demand services and online content, is expected to contribute to a better quality of experience for a costumer of such a service. However, the lack of efficient methods for finding the right content, adapted to personal interests, may lead to a progressive loss of clients. In such a scenario, recommendation systems are seen as a tool that can fill this gap and contribute to the loyalty of users. Multimedia content, namely films and television programmes are usually described using a set of metadata elements that include the title, a genre, the date of production, and the list of directors and actors. This paper provides a deep study on how the use of different metadata elements can contribute to increase the quality of the recommendations suggested. The analysis is conducted using Netflix and Movielens datasets and aspects such as the granularity of the descriptions, the accuracy metric used and the sparsity of the data are taken into account. Comparisons with collaborative approaches are also presented.

关键词： recommendation algorithms Collaborative Content-based Metadata

来源：评论

学校读者我要写书评

暂无评论

KMPR-AEP: Knowledge-Enhanced Multi-task Parallelized recommendation Algorithm Incorporating Attention-Embedded Propagation

引用

INTERNATIONAL JOURNAL OF COMPUTATIONAL INTELLIGENCE SYSTEMS 2024年第1期17卷 1-21页

作者： Zhang, Yang Cai, Juanjuan Li, Chuanzhen Li, Tong Wang, Hui Commun Univ China Sch Informat & Commun Engn Beijing 100024 Peoples R China Commun Univ China Key Lab Media Audio & Video Minist Educ Beijing 100024 Peoples R China Commun Univ China State Key Lab Media Convergence & Commun Beijing 100024 Peoples R China

Existing knowledge graph-based recommendation algorithms either use Knowledge Graphs (KGs) as auxiliary information or use KG prediction tasks as regular terms to constrain recommendation with multi-task learning. However, the first method, which introduces multi-hop neighbors to enhance item representations, also weakens the relationship information within individual triples to some extent. The second method ignores neighboring information and thus fails to capture the long-range connectivity between items, which leads to a lack of utilization of the structural information in the KG. In response to the limitations of existing recommendation algorithms that do not fully leverage the relational and structural information within KGs, we propose a novel method named Knowledge-Enhanced Multi-Task Parallelized recommendation Algorithm Incorporating Attention-Embedded Propagation (KMPR-AEP). It employs a parallel approach to simultaneously incorporate KG as auxiliary information and a regular term. We take into account the multi-hop neighbor information of users and items as well as enhance the relations among individual triples, which considers both the structural and relational information of KG. Extensive experiments conducted on three real-world datasets for CTR prediction and Top-K recommendation scenarios demonstrate the superiority of our proposed method over the state-of-the-art. On average, our KMPR-AEP shows improvements of 2.44% in AUC, 3.49% in ACC, and 2.64% in F1.

关键词： Attention embedding propagation Knowledge graph embedding recommendation algorithms Multi-task

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：