scholarly journals Cross-Domain Recommendation: Challenges, Progress, and Prospects

2021 ◽  
Author(s):  
Feng Zhu ◽  
Yan Wang ◽  
Chaochao Chen ◽  
Jun Zhou ◽  
Longfei Li ◽  
...  
Keyword(s):  
Systematic Review ◽  
Recommender Systems ◽  
Research Progress ◽  
Data Sparsity ◽  
Comprehensive Review ◽  
Full View ◽  
Cross Domain ◽  
Single Target ◽  
Sparsity Problem ◽  
Dual Target

To address the long-standing data sparsity problem in recommender systems (RSs), cross-domain recommendation (CDR) has been proposed to leverage the relatively richer information from a richer domain to improve the recommendation performance in a sparser domain. Although CDR has been extensively studied in recent years, there is a lack of a systematic review of the existing CDR approaches. To fill this gap, in this paper, we provide a comprehensive review of existing CDR approaches, including challenges, research progress, and prospects. Specifically, we first summarize existing CDR approaches into four types, including single-target CDR, single-target multi-domain recommendation (MDR), dual-target CDR, and multi-target CDR. We then present the definitions and challenges of these CDR approaches. Next, we propose a full-view categorization and new taxonomies on these approaches and report their research progress in detail. In the end, we share several promising prospects in CDR.

Combining Item Rating Similarity and Item Classification Similarity for Better Recommendation Quality

2012 ◽  
Vol 461 ◽  
pp. 289-292
Author(s):  
Kai Zhou
Keyword(s):  
Collaborative Filtering ◽  
Recommender Systems ◽  
Fundamental Problem ◽  
The Other ◽  
Recommendation Algorithm ◽  
Data Sparsity ◽  
Filtering Method ◽  
Product Recommendation ◽  
Sparsity Problem

Recommender systems are becoming increasingly popular, and collaborative filtering method is one of the most important technologies in recommender systems. The ability of recommender systems to make correct predictions is fundamentally determined by the quality and fittingness of the collaborative filtering that implements them. It is currently mainly used for business purposes such as product recommendation. Collaborative filtering has two types. One is user based collaborative filtering using the similarity between users to predict and the other is item based collaborative filtering using the similarity between items. Although both of them are successfully applied in wide regions, they suffer from a fundamental problem of data sparsity. This paper gives a personalized collaborative filtering recommendation algorithm combining the item rating similarity and the item classification similarity. This method can alleviate the data sparsity problem in the recommender systems

An Analysis of Group Recommendation Strategies

2010 ◽  
Vol 14 (6) ◽  
pp. 729-734 ◽  
Author(s):  
Shlomo Berkovsky ◽  
◽  
Jill Freyne
Keyword(s):  
Collaborative Filtering ◽  
Recommender Systems ◽  
Insufficient Data ◽  
Group Recommendation ◽  
Data Sparsity ◽  
Aggregated Data ◽  
Sparsity Problem ◽  
Group Recommendations

Collaborative filtering recommender systems often suffer from a data sparsity problem, where systems have insufficient data to generate accurate recommendations. To partially resolve this, the use of group aggregated data in the collaborative filtering recommendations process has been suggested. Although group recommendations are typically less accurate than personalized recommendations, they can be more accurate than generic ones, which are the natural fall back when personalized recommendations cannot be generated. This work presents a study that exploits a dataset of recipe ratings from families of users, in order to evaluate the accuracy of several group recommendation strategies and weighting models.

Alleviating the data sparsity problem of recommender systems by clustering nodes in bipartite networks

2020 ◽  
Vol 149 ◽  
pp. 113346
Author(s):  
Fuguo Zhang ◽  
Shumei Qi ◽  
Qihua Liu ◽  
Mingsong Mao ◽  
An Zeng
Keyword(s):  
Recommender Systems ◽  
Bipartite Networks ◽  
Data Sparsity ◽  
Sparsity Problem

scholarly journals A Graphical and Attentional Framework for Dual-Target Cross-Domain Recommendation

2020 ◽  
Author(s):  
Feng Zhu ◽  
Yan Wang ◽  
Chaochao Chen ◽  
Guanfeng Liu ◽  
Xiaolin Zheng
Keyword(s):  
State Of The Art ◽  
The Novel ◽  
Target Domain ◽  
Cross Domain ◽  
Single Target ◽  
Recommendation Accuracy ◽  
Real World Datasets ◽  
Content Information ◽  
Dual Target

The conventional single-target Cross-Domain Recommendation (CDR) only improves the recommendation accuracy on a target domain with the help of a source domain (with relatively richer information). In contrast, the novel dual-target CDR has been proposed to improve the recommendation accuracies on both domains simultaneously. However, dual-target CDR faces two new challenges: (1) how to generate more representative user and item embeddings, and (2) how to effectively optimize the user/item embeddings on each domain. To address these challenges, in this paper, we propose a graphical and attentional framework, called GA-DTCDR. In GA-DTCDR, we first construct two separate heterogeneous graphs based on the rating and content information from two domains to generate more representative user and item embeddings. Then, we propose an element-wise attention mechanism to effectively combine the embeddings of common users learned from both domains. Both steps significantly enhance the quality of user and item embeddings and thus improve the recommendation accuracy on each domain. Extensive experiments conducted on four real-world datasets demonstrate that GA-DTCDR significantly outperforms the state-of-the-art approaches.

scholarly journals Knowledge Transfer via Pre-training for Recommendation: A Review and Prospect

2021 ◽  
Vol 4 ◽  
Author(s):  
Zheni Zeng ◽  
Chaojun Xiao ◽  
Yuan Yao ◽  
Ruobing Xie ◽  
Zhiyuan Liu ◽  
...  
Keyword(s):  
Knowledge Transfer ◽  
Recommender Systems ◽  
Real World ◽  
Source Code ◽  
Cold Start ◽  
Future Research ◽  
Data Sparsity ◽  
Sparsity Problem

Recommender systems aim to provide item recommendations for users and are usually faced with data sparsity problems (e.g., cold start) in real-world scenarios. Recently pre-trained models have shown their effectiveness in knowledge transfer between domains and tasks, which can potentially alleviate the data sparsity problem in recommender systems. In this survey, we first provide a review of recommender systems with pre-training. In addition, we show the benefits of pre-training to recommender systems through experiments. Finally, we discuss several promising directions for future research of recommender systems with pre-training. The source code of our experiments will be available to facilitate future research.

Dl-3-n-Butylphthalide (NBP): A Promising Therapeutic Agent for Ischemic Stroke

2018 ◽  
Vol 17 (5) ◽  
pp. 338-347 ◽  
Author(s):  
Shan Wang ◽  
Fei Ma ◽  
Longjian Huang ◽  
Yong Zhang ◽  
Yuchen Peng ◽  
...  
Keyword(s):  
Ischemic Stroke ◽  
Complex Disease ◽  
Time Window ◽  
Recombinant Tissue Plasminogen Activator ◽  
Research Progress ◽  
Apium Graveolens ◽  
Synthetic Compound ◽  
Stroke Treatment ◽  
Single Target ◽  
Anti Oxidant

Background and Objective: Stroke is a leading cause of morbidity and mortality in both developed and developing countries all over the world. The only drug for ischemic stroke approved by FDA is recombinant tissue plasminogen activator (rtPA). However, only 2-5% stroke patients receive rtPAs treatment due to its strict therapeutic time window. As ischemic stroke is a complex disease involving multiple mechanisms, medications with multi-targets may be more powerful compared with single-target drugs. Dl-3-n-Butylphthalide (NBP) is a synthetic compound based on l-3-n- Butylphthalide that is isolated from seeds of Apium graveolens. The racemic 3-n-butylphthalide (dl- NBP) was approved by Food and Drug Administration of China for the treatment of ischemic stroke in 2002. A number of clinical studies indicated that NBP not only improved the symptoms of ischemic stroke, but also contributed to the long-term recovery. The potential mechanisms of NBP for ischemic stroke treatment may target different pathophysiological processes, including anti-oxidant, antiinflammation, anti-apoptosis, anti-thrombosis, and protection of mitochondria et al. Conclusion: In this review, we have summarized the research progress of NBP for the treatment of ischemic stroke during the past two decades.

scholarly journals Research on the Operating Characteristics of Hydraulic Free-Piston Engines: A Systematic Review and Meta-Analysis

Energies ◽  
2021 ◽  
Vol 14 (12) ◽  
pp. 3530
Author(s):  
Fukang Ma ◽  
Shuanlu Zhang ◽  
Zhenfeng Zhao ◽  
Yifang Wang
Keyword(s):  
Systematic Review ◽  
High Efficiency ◽  
Meta Analysis ◽  
Combustion Process ◽  
Research Progress ◽  
Future Research ◽  
Operating Characteristics ◽  
Piston Engine ◽  
Free Piston ◽  
Free Piston Engine

The hydraulic free-piston engine (HFPE) is a kind of hybrid-powered machine which combines the reciprocating piston-type internal combustion engine and the plunger pump as a whole. In recent years, the HFPE has been investigated by a number of research groups worldwide due to its potential advantages of high efficiency, energy savings, reduced emissions and multi-fuel operation. Therefore, our study aimed to assess the operating characteristics, core questions and research progress of HFPEs via a systematic review and meta-analysis. We included operational control, starting characteristics, misfire characteristics, in-cylinder working processes and operating stability. We conducted the literature search using electronic databases. The research on HFPEs has mainly concentrated on four kinds of free-piston engine, according to piston arrangement form: single piston, dual pistons, opposed pistons and four-cylinder complex configuration. HFPE research in China is mainly conducted in Zhejiang University, Tianjin University, Jilin University and the Beijing Institute of Technology. In addition, in China, research has mainly focused on the in-cylinder combustion process while a piston is free by considering in-cylinder combustion machinery and piston dynamics. Regarding future research, it is very important that we solve the instabilities brought about by chance fluctuations in the combustion process, which will involve the hydraulic system’s efficiency, the cyclical variation, the method of predicting instability and the recovery after instability.

scholarly journals Efficacy and safety of neoadjuvant therapy for HER2-positive early breast cancer: a network meta-analysis

2021 ◽  
Vol 13 ◽  
pp. 175883592110069
Author(s):  
Jie Zhang ◽  
Yushuai Yu ◽  
Yuxiang Lin ◽  
Shaohong Kang ◽  
Xinyin Lv ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Neoadjuvant Therapy ◽  
Combination Chemotherapy ◽  
Target Therapy ◽  
Meta Analysis ◽  
Her2 Positive ◽  
Single Target ◽  
Dual Target ◽  
Better Than ◽  
Positive Breast Cancer

Aims: Currently, there are many approaches available for neoadjuvant therapy for human epidermal growth factor receptor 2 (HER2)-positive breast cancer that improve therapeutic efficacy but are also controversial. We conducted a two-step Bayesian network meta-analysis (NMA) to compare odds ratios (ORs) for pathologic complete response (PCR) and safety endpoints. Methods: The Cochrane Central Register of Controlled Trials, PubMed, Embase, and online abstracts from the American Society of Clinical Oncology and San Antonio Breast Cancer Symposium were searched comprehensively and systematically. Phase II/III randomised clinical trials for targeted therapy in at least one arm were included. Results: A total of 9779 published manuscripts were identified, and 36 studies including 10,379 patients were finally included in our analysis. The NMA of PCR showed that dual-target therapy is better than single-target therapy and combination chemotherapy is better than monochemotherapy. However, anthracycline did not bring extra benefits, whether combined with dual-target therapy or single-target therapy. On the other hand, the addition of endocrine therapy in the HER2-positive, hormone receptor (HR)-positive subgroup might have additional beneficial effects but without significant statistical difference. By performing a conjoint analysis of the PCR rate and safety endpoints, we found that ‘trastuzumab plus pertuzumab’ and ‘T-DM1 containing regimens’ were well balanced in terms of efficacy and toxicity in all target regimens. Conclusion: In summary, trastuzumab plus pertuzumab-based dual-target therapy with combination chemotherapy regimens showed the highest efficacy of all optional regimens. They also achieved the best balance between efficacy and toxicity. As our study showed that anthracycline could be replaced by carboplatin, we strongly recommended TCbHP as the preferred choice for neoadjuvant treatment of HER2-positive breast cancer. We also look forward to the potential value of T-DM1 in improving outcomes, which needs further study in future trials.

Sign in / Sign up

Export Citation Format

Share Document