scholarly journals HYBRID RECOMMENDATION ALGORITHM BASED ON TWO ROLES OF SOCIAL TAGS

2012 ◽  
Vol 22 (07) ◽  
pp. 1250166 ◽  
Author(s):  
ZI-KE ZHANG ◽  
CHUANG LIU
Keyword(s):  
Hybrid Algorithm ◽  
Social Tagging ◽  
Great Success ◽  
Real World Data ◽  
Recommendation Algorithm ◽  
Data Set ◽  
The Past ◽  
Social Tags ◽  
New Family ◽  
Hybrid Recommendation

The past few years have witnessed the great success of a new family of paradigms, social tagging networks, which allows users to freely associate social tags to items and efficiently manage them. Thus it provides us a promising way to effectively find useful and interesting information. In this paper, we consider two typical roles of social tags: (i) an accessorial tool helping users organize items; (ii) a bridge that connects users and items. We then propose a hybrid algorithm to integrate the two different roles to obtain better recommendation performance. Experimental results on a real-world data set, Del.icio.us, shows that it can significantly enhance both the algorithmic accuracy and diversity.

Social Tagging

2010 ◽  
pp. 260-276
Author(s):  
Yong Yu ◽  
Rui Li ◽  
Shenghua Bao ◽  
Ben Fei ◽  
Zhong Su
Keyword(s):  
Web Sites ◽  
Social Tagging ◽  
Great Success ◽  
Power Law Distribution ◽  
Web Resources ◽  
Social Tags ◽  
The Social ◽  
Social Annotations ◽  
Quality Property

Recently, collaborative tagging Web sites such as Del.icio.us and Flickr have achieved great success. This chapter is concerned with the problem of social tagging analysis and mining. More specifically, we discuss five properties of social tagging and their applications: 1) keyword property, which means social annotations serve as human selected keywords for Web resources; 2) semantic property, which indicates semantic relations among tags and Web resources; 3) hierarchical property, which means that hierarchical structure can be derived from the flat social tagging space; 4) quality property, which means that Web resources’ qualities are varied and can be quantified using social tagging; 5) distribution property, which indicates the distribution of frequencies of social tags usually converges to a power-law distribution. These properties are the most principle characteristics, which have been popularly discussed and explored in many applications. As a case study, we show how to improve the social resource browsing by applying the five properties of social tags.

A Service Recommendation Algorithm Based on Modeling of Dynamic and Diverse Demands

2018 ◽  
Vol 15 (1) ◽  
pp. 47-70 ◽  
Author(s):  
Yanmei Zhang ◽  
Tingpei Lei ◽  
Zhiguang Qin
Keyword(s):  
Latent Dirichlet Allocation ◽  
Service Recommendation ◽  
Real World Data ◽  
Recommendation Algorithm ◽  
Data Set ◽  
Allocation Model ◽  
Latent Dirichlet Allocation Model ◽  
Graph Modeling ◽  
Recommendation Algorithms ◽  
User Demand

This article contends that current service recommendation algorithms are still unable to meet the dynamic and diverse demands of users, so a service recommendation algorithm considering dynamic and diverse demands is proposed. The latent Dirichlet allocation model of machine learning field is adopted to extract the user implicit demand factors, and then the bipartite graph modeling and random-walk algorithm are used to extend implicit demand factors to predict short-term changes and diversity of user demand. At last, the service recommendation list is generated based on these demand factors. Experimental results on a real-world data set regarding service composition show that the proposed algorithm can represent dynamic and diverse user demands, and the performance of the proposed algorithm is better than that of the other algorithms in terms of accuracy, novelty, and diversity.

scholarly journals Multi-Label Causal Feature Selection

2020 ◽  
Vol 34 (04) ◽  
pp. 6430-6437 ◽  
Author(s):  
Xingyu Wu ◽  
Bingbing Jiang ◽  
Kui Yu ◽  
Huanhuan Chen ◽  
Chunyan Miao
Keyword(s):  
Feature Selection ◽  
State Of The Art ◽  
Causal Mechanism ◽  
Data Sets ◽  
Causal Relationships ◽  
Real World Data ◽  
Data Set ◽  
Complete Representation ◽  
The Past ◽  
Different Types

Multi-label feature selection has received considerable attentions during the past decade. However, existing algorithms do not attempt to uncover the underlying causal mechanism, and individually solve different types of variable relationships, ignoring the mutual effects between them. Furthermore, these algorithms lack of interpretability, which can only select features for all labels, but cannot explain the correlation between a selected feature and a certain label. To address these problems, in this paper, we theoretically study the causal relationships in multi-label data, and propose a novel Markov blanket based multi-label causal feature selection (MB-MCF) algorithm. MB-MCF mines the causal mechanism of labels and features first, to obtain a complete representation of information about labels. Based on the causal relationships, MB-MCF then selects predictive features and simultaneously distinguishes common features shared by multiple labels and label-specific features owned by single labels. Experiments on real-world data sets validate that MB-MCF could automatically determine the number of selected features and simultaneously achieve the best performance compared with state-of-the-art methods. An experiment in Emotions data set further demonstrates the interpretability of MB-MCF.

TrustSVD: A Novel Trust-Based Matrix Factorization Model with User Trust and Item Ratings

2017 ◽  
Vol 7 (11) ◽  
pp. 7 ◽  
Author(s):  
K Sobha Rani
Keyword(s):  
Matrix Factorization ◽  
Social Trust ◽  
State Of The Art ◽  
Data Sets ◽  
Real World Data ◽  
Recommendation Algorithm ◽  
Active User ◽  
Factorization Model ◽  
The Social ◽  
Matrix Factorization Technique

Collaborative filtering suffers from the problems of data sparsity and cold start, which dramatically degrade recommendation performance. To help resolve these issues, we propose TrustSVD, a trust-based matrix factorization technique. By analyzing the social trust data from four real-world data sets, we conclude that not only the explicit but also the implicit influence of both ratings and trust should be taken into consideration in a recommendation model. Hence, we build on top of a state-of-the-art recommendation algorithm SVD++ which inherently involves the explicit and implicit influence of rated items, by further incorporating both the explicit and implicit influence of trusted users on the prediction of items for an active user. To our knowledge, the work reported is the first to extend SVD++ with social trust information. Experimental results on the four data sets demonstrate that our approach TrustSVD achieves better accuracy than other ten counterparts, and can better handle the concerned issues.

scholarly journals Unprecedented Coordination-Induced Bright Red Emission from Group 12 Metal-Bound Triarylazoimidazoles

Molecules ◽  
2021 ◽  
Vol 26 (6) ◽  
pp. 1739
Author(s):  
Artyom A. Astafiev ◽  
Olga V. Repina ◽  
Boris S. Tupertsev ◽  
Alexey A. Nazarov ◽  
Maria R. Gonchar ◽  
...  
Keyword(s):  
Visible Light ◽  
Metal Complexes ◽  
Scientific Literature ◽  
Photo Luminescence ◽  
Red Emission ◽  
The Past ◽  
Highly Efficient ◽  
Luminescence Emission ◽  
New Family ◽  
Group 12

Arylazoimidazoles are important dyes which were intensively studied in the past. In contrast, triarylazoimidazoles (derivatives which carry aryl substituents at the imidazole core) received almost no attention in the scientific literature. Here, we report a new family of simple and easily accessible triarylazoimidazole-group 12 metal complexes, which feature highly efficient photo-luminescence emission (Φ up to  0.44). Novel compounds exhibit bright red emission in solution, which could be excited with a visible light.

Dynamic network intelligent hybrid recommendation algorithm and its application in online shopping platform

2021 ◽  
pp. 1-13
Author(s):  
Yuxuan Gao ◽  
Haiming Liang ◽  
Bingzhen Sun
Keyword(s):  
Online Shopping ◽  
Rapid Development ◽  
Dynamic Network ◽  
Topsis Method ◽  
Recommendation Algorithm ◽  
Time Periods ◽  
The Difference ◽  
Relationship Of ◽  
Hybrid Recommendation ◽  
Novel Model

With the rapid development of e-commerce, whether network intelligent recommendation can attract customers has become a measure of customer retention on online shopping platforms. In the literature about network intelligent recommendation, there are few studies that consider the difference preference of customers in different time periods. This paper proposes the dynamic network intelligent hybrid recommendation algorithm distinguishing time periods (DIHR), it is a integrated novel model combined with the DEMATEL and TOPSIS method to solved the problem of network intelligent recommendation considering time periods. The proposed method makes use of the DEMATEL method for evaluating the preference relationship of customers for indexes of merchandises, and adopt the TOPSIS method combined with intuitionistic fuzzy number (IFN) for assessing and ranking the merchandises according to the indexes. We specifically introduce the calculation steps of the proposed method, and then calculate its application in the online shopping platform.

scholarly journals Change in mean and extreme temperature at Yingkou station in Northeast China from 1904 to 2017

2021 ◽  
Vol 164 (3-4) ◽  
Author(s):  
Xiaoying Xue ◽  
Guoyu Ren ◽  
Xiubao Sun ◽  
Panfeng Zhang ◽  
Yuyu Ren ◽  
...  
Keyword(s):  
Northeast China ◽  
Extreme Temperature ◽  
Maximum Temperature ◽  
Data Set ◽  
The Past ◽  
Significant Downward Trend ◽  
Cold Events ◽  
Temperature Indices ◽  
Increasing Trends ◽  
Change In Mean

AbstractThe understanding of centennial trends of extreme temperature has been impeded due to the lack of early-year observations. In this paper, we collect and digitize the daily temperature data set of Northeast China Yingkou meteorological station since 1904. After quality control and homogenization, we analyze the changes of mean and extreme temperature in the past 114 years. The results show that mean temperature (Tmean), maximum temperature (Tmax), and minimum temperature (Tmin) all have increasing trends during 1904–2017. The increase of Tmin is the most obvious with the rate of 0.34 °C/decade. The most significant warming occurs in spring and winter with the rate of Tmean reaching 0.32 °C/decade and 0.31 °C/decade, respectively. Most of the extreme temperature indices as defined using absolute and relative thresholds of Tmax and Tmin also show significant changes, with cold events witnessing a more significant downward trend. The change is similar to that reported for global land and China for the past six decades. It is also found that the extreme highest temperature (1958) and lowest temperature (1920) records all occurred in the first half of the whole period, and the change of extreme temperature indices before 1950 is different from that of the recent decades, in particular for diurnal temperature range (DTR), which shows an opposite trend in the two time periods.

Auto-sharing parameters for transfer learning based on multi-objective optimization

2021 ◽  
pp. 1-13
Author(s):  
Hailin Liu ◽  
Fangqing Gu ◽  
Zixian Lin
Keyword(s):  
Transfer Learning ◽  
Optimization Problem ◽  
Data Sets ◽  
Multi Objective Optimization ◽  
Particle Swarm Optimizer ◽  
Real World Data ◽  
Data Set ◽  
Target Task ◽  
Main Research ◽  
Multi Objective

Transfer learning methods exploit similarities between different datasets to improve the performance of the target task by transferring knowledge from source tasks to the target task. “What to transfer” is a main research issue in transfer learning. The existing transfer learning method generally needs to acquire the shared parameters by integrating human knowledge. However, in many real applications, an understanding of which parameters can be shared is unknown beforehand. Transfer learning model is essentially a special multi-objective optimization problem. Consequently, this paper proposes a novel auto-sharing parameter technique for transfer learning based on multi-objective optimization and solves the optimization problem by using a multi-swarm particle swarm optimizer. Each task objective is simultaneously optimized by a sub-swarm. The current best particle from the sub-swarm of the target task is used to guide the search of particles of the source tasks and vice versa. The target task and source task are jointly solved by sharing the information of the best particle, which works as an inductive bias. Experiments are carried out to evaluate the proposed algorithm on several synthetic data sets and two real-world data sets of a school data set and a landmine data set, which show that the proposed algorithm is effective.

Sign in / Sign up

Export Citation Format

Share Document