Learning a Hierarchical Intent Model for Next-Item Recommendation

2022 ◽  
Vol 40 (2) ◽  
pp. 1-28
Author(s):  
Nengjun Zhu ◽  
Jian Cao ◽  
Xinjiang Lu ◽  
Hui Xiong
Keyword(s):  
Contextual Information ◽  
User Profile ◽  
User Preference ◽  
User Profiles ◽  
Factors Affecting ◽  
User Intent ◽  
Semantic Component ◽  
Two Factors ◽  
Memory Network ◽  
Real World Datasets

A session-based recommender system (SBRS) captures users’ evolving behaviors and recommends the next item by profiling users in terms of items in a session. User intent and user preference are two factors affecting his (her) decisions. Specifically, the former narrows the selection scope to some item types, while the latter helps to compare items of the same type. Most SBRSs assume one arbitrary user intent dominates a session when making a recommendation. However, this oversimplifies the reality that a session may involve multiple types of items conforming to different intents. In current SBRSs, items conforming to different user intents have cross-interference in profiling users for whom only one user intent is considered. Explicitly identifying and differentiating items conforming to various user intents can address this issue and model rich contextual information of a session. To this end, we design a framework modeling user intent and preference explicitly, which empowers the two factors to play their distinctive roles. Accordingly, we propose a key-array memory network (KA-MemNN) with a hierarchical intent tree to model coarse-to-fine user intents. The two-layer weighting unit (TLWU) in KA-MemNN detects user intents and generates intent-specific user profiles. Furthermore, the hierarchical semantic component (HSC) integrates multiple sets of intent-specific user profiles along with different user intent distributions to model a multi-intent user profile. The experimental results on real-world datasets demonstrate the superiority of KA-MemNN over selected state-of-the-art methods.

scholarly journals NeuSE: A Neural Snapshot Ensemble Method for Collaborative Filtering

2021 ◽  
Vol 15 (6) ◽  
pp. 1-20
Author(s):  
Dongsheng Li ◽  
Haodong Liu ◽  
Chao Chen ◽  
Yingying Zhao ◽  
Stephen M. Chu ◽  
...  
Keyword(s):  
Collaborative Filtering ◽  
Optimization Problems ◽  
Empirical Studies ◽  
Large Datasets ◽  
Model Learning ◽  
Global Models ◽  
Convex Optimization Problems ◽  
Memory Network ◽  
Real World Datasets ◽  
Performance Tradeoff

In collaborative filtering (CF) algorithms, the optimal models are usually learned by globally minimizing the empirical risks averaged over all the observed data. However, the global models are often obtained via a performance tradeoff among users/items, i.e., not all users/items are perfectly fitted by the global models due to the hard non-convex optimization problems in CF algorithms. Ensemble learning can address this issue by learning multiple diverse models but usually suffer from efficiency issue on large datasets or complex algorithms. In this article, we keep the intermediate models obtained during global model learning as the snapshot models, and then adaptively combine the snapshot models for individual user-item pairs using a memory network-based method. Empirical studies on three real-world datasets show that the proposed method can extensively and significantly improve the accuracy (up to 15.9% relatively) when applied to a variety of existing collaborative filtering methods.

Initial weight and virus dose: two factors affecting the onset of acute coxsackievirus B3 myocarditis in C57BL/6 mouse—a histopathology-based study

2013 ◽  
Vol 22 (1) ◽  
pp. 96-101 ◽  
Author(s):  
Minghui Li ◽  
Xinggang Wang ◽  
Yuquan Xie ◽  
Yeqing Xie ◽  
Xian Zhang ◽  
...  
Keyword(s):  
Coxsackievirus B3 ◽  
Initial Weight ◽  
Factors Affecting ◽  
Virus Dose ◽  
Two Factors

Characterising Usage of a Multicomponent Digital Intervention to Predict Treatment Outcomes in First-Episode Psychosis: A Cluster Analysis (Preprint)

2021 ◽  
Author(s):  
Shaunagh O'Sullivan ◽  
Lianne Schmaal ◽  
Simon D'Alfonso ◽  
Yara J Toenders ◽  
Lee Valentine ◽  
...  
Keyword(s):  
Young People ◽  
Treatment Outcomes ◽  
Social Functioning ◽  
Negative Symptoms ◽  
User Profile ◽  
First Episode Psychosis ◽  
First Episode ◽  
User Profiles ◽  
Episode Psychosis ◽  
Social Components

BACKGROUND Multicomponent digital interventions offer the potential for tailored and flexible interventions that aim to address high attrition rates and increase engagement, an area of concern in digital mental health. However, increased flexibility in usage makes it difficult to determine which components lead to improved treatment outcomes. OBJECTIVE This study aimed to identify user profiles on Horyzons, an 18-month digital relapse prevention intervention that incorporates therapeutic content and social networking, along with clinical, vocational and peer support, and to examine the predictive value of these user profiles for treatment outcomes. A secondary objective was to compare each user profile with young people receiving treatment as usual (TAU). METHODS Participants comprised 82 young people (16-27 years of age) with access to Horyzons and 84 receiving TAU, recovering from first-episode psychosis. Six-month usage data from the therapy and social networking components of Horyzons were used as features for K-means clustering for joint trajectories to identify user profiles. Social functioning, psychotic symptoms, depression and anxiety were assessed at baseline and six-month follow-up. General linear mixed models were used to examine the predictive value of user profiles for treatment outcomes, and between each user profile with TAU. RESULTS Three user profiles were identified based on system usage metrics including: (a) low usage; (b) maintained usage of social components; and (c) maintained usage of both therapy and social components. The maintained therapy and social group showed improvements in social functioning (F (2, 51) = 3.58; P = .04), negative symptoms (F (2, 51) = 4.45; P = .02) and overall psychiatric symptom severity (F (2, 50) = 3.23; P = .048) compared to the other user profiles. This group also showed improvements in social functioning (F (1, 62) = 4.68; P = .03), negative symptoms (F (1, 62) = 14.61; P = <.001) and overall psychiatric symptom severity (F (1, 63) = 5.66; P = .02) compared to TAU. Conversely, the maintained social group showed increases in anxiety compared to TAU (F (1, 57) = 7.65; P = .01). No differences were found between the low usage group and TAU on treatment outcomes. CONCLUSIONS Continued engagement with both therapy and social components might be key in achieving long-term recovery. Maintained social usage and low usage outcomes were broadly comparable to TAU, emphasizing the importance of maintaining engagement for improved treatment outcomes. Although the social network may be a key ingredient to increase sustained engagement, as users engaged with this more consistently, it should be leveraged as a tool to engage young people with therapeutic content to bring about social and clinical benefits.

An Experimental Study of Spammer Detection on Chinese Microblogs

2020 ◽  
Vol 30 (11n12) ◽  
pp. 1759-1777
Author(s):  
Jialing Liang ◽  
Peiquan Jin ◽  
Lin Mu ◽  
Jie Zhao
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
User Behavior ◽  
Real Data ◽  
User Profile ◽  
Data Set ◽  
Sina Weibo ◽  
Factors Affecting ◽  
The Government ◽  
Hot Event

With the development of Web 2.0, social media such as Twitter and Sina Weibo have become an essential platform for disseminating hot events. Simultaneously, due to the free policy of microblogging services, users can post user-generated content freely on microblogging platforms. Accordingly, more and more hot events on microblogging platforms have been labeled as spammers. Spammers will not only hurt the healthy development of social media but also introduce many economic and social problems. Therefore, the government and enterprises must distinguish whether a hot event on microblogging platforms is a spammer or is a naturally-developing event. In this paper, we focus on the hot event list on Sina Weibo and collect the relevant microblogs of each hot event to study the detecting methods of spammers. Notably, we develop an integral feature set consisting of user profile, user behavior, and user relationships to reflect various factors affecting the detection of spammers. Then, we employ typical machine learning methods to conduct extensive experiments on detecting spammers. We use a real data set crawled from the most prominent Chinese microblogging platform, Sina Weibo, and evaluate the performance of 10 machine learning models with five sampling methods. The results in terms of various metrics show that the Random Forest model and the over-sampling method achieve the best accuracy in detecting spammers and non-spammers.

scholarly journals Using a Novel Grey DANP Model to Identify Interactions between Manufacturing and Logistics Industries in China

2018 ◽  
Vol 10 (10) ◽  
pp. 3456 ◽  
Author(s):  
Peng Jiang ◽  
Yi-Chung Hu ◽  
Ghi-Feng Yen ◽  
Hang Jiang ◽  
Yu-Jing Chiu
Keyword(s):  
Manufacturing Industry ◽  
Multiple Criteria Decision Making ◽  
Value Added ◽  
Manufacturing Industries ◽  
Direct Influence ◽  
Key Factors ◽  
Logistics Industry ◽  
Influence Matrix ◽  
Factors Affecting ◽  
Two Factors

As a crucial part of producer services, the logistics industry is highly dependent on the manufacturing industry. In general, the interactive development of the logistics and manufacturing industries is essential. Due to the existence of a certain degree of interdependence between any two factors, interaction between the two industries has produced a basis for measurement; identifying the key factors affecting the interaction between the manufacturing and logistics industries is a kind of decision problem in the field of multiple criteria decision making (MCDM). A hybrid MCDM method, DEMATEL-based ANP (DANP) is appropriate to solve this problem. However, DANP uses a direct influence matrix, which involves pairwise comparisons that may be more or less influenced by the respondents. Therefore, we propose a decision model, Grey DANP, which can automatically generate the direct influence matrix. Statistical data for the logistics and manufacturing industries in the China Statistical Yearbook (2006–2015) were used to identify the key factors for interaction between these two industries. The results showed that the key logistics criteria for interaction development are the total number of employees in the transport business, the volume of goods, and the total length of routes. The key manufacturing criteria for interaction development are the gross domestic product and the value added. Therefore, stakeholders should increase the number of employees in the transport industry and freight volumes. Also, the investment in infrastructure should be increased.

A study of user profile representation for personalized cross-language information retrieval

2016 ◽  
Vol 68 (4) ◽  
pp. 448-477 ◽  
Author(s):  
Dong Zhou ◽  
Séamus Lawless ◽  
Xuan Wu ◽  
Wenyu Zhao ◽  
Jianxun Liu
Keyword(s):  
Information Retrieval ◽  
Query Expansion ◽  
User Profile ◽  
User Profiles ◽  
Content Type ◽  
Cross Language Information Retrieval ◽  
Cross Lingual ◽  
Cross Language ◽  
Representation Techniques ◽  
Comprehensive Study

Purpose – With an increase in the amount of multilingual content on the World Wide Web, users are often striving to access information provided in a language of which they are non-native speakers. The purpose of this paper is to present a comprehensive study of user profile representation techniques and investigate their use in personalized cross-language information retrieval (CLIR) systems through the means of personalized query expansion. Design/methodology/approach – The user profiles consist of weighted terms computed by using frequency-based methods such as tf-idf and BM25, as well as various latent semantic models trained on monolingual documents and cross-lingual comparable documents. This paper also proposes an automatic evaluation method for comparing various user profile generation techniques and query expansion methods. Findings – Experimental results suggest that latent semantic-weighted user profile representation techniques are superior to frequency-based methods, and are particularly suitable for users with a sufficient amount of historical data. The study also confirmed that user profiles represented by latent semantic models trained on a cross-lingual level gained better performance than the models trained on a monolingual level. Originality/value – Previous studies on personalized information retrieval systems have primarily investigated user profiles and personalization strategies on a monolingual level. The effect of utilizing such monolingual profiles for personalized CLIR remains unclear. The current study fills the gap by a comprehensive study of user profile representation for personalized CLIR and a novel personalized CLIR evaluation methodology to ensure repeatable and controlled experiments can be conducted.

scholarly journals Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze

Electronics ◽  
2021 ◽  
Vol 10 (22) ◽  
pp. 2868
Author(s):  
Wenxuan Zhao ◽  
Yaqin Zhao ◽  
Liqi Feng ◽  
Jiaxi Tang
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Real World ◽  
Large Scale ◽  
Learning Strategy ◽  
Contextual Information ◽  
Small Scale ◽  
Image Dehazing ◽  
Atmospheric Scattering ◽  
Real World Datasets

The purpose of image dehazing is the reduction of the image degradation caused by suspended particles for supporting high-level visual tasks. Besides the atmospheric scattering model, convolutional neural network (CNN) has been used for image dehazing. However, the existing image dehazing algorithms are limited in face of unevenly distributed haze and dense haze in real-world scenes. In this paper, we propose a novel end-to-end convolutional neural network called attention enhanced serial Unet++ dehazing network (AESUnet) for single image dehazing. We attempt to build a serial Unet++ structure that adopts a serial strategy of two pruned Unet++ blocks based on residual connection. Compared with the simple Encoder–Decoder structure, the serial Unet++ module can better use the features extracted by encoders and promote contextual information fusion in different resolutions. In addition, we take some improvement measures to the Unet++ module, such as pruning, introducing the convolutional module with ResNet structure, and a residual learning strategy. Thus, the serial Unet++ module can generate more realistic images with less color distortion. Furthermore, following the serial Unet++ blocks, an attention mechanism is introduced to pay different attention to haze regions with different concentrations by learning weights in the spatial domain and channel domain. Experiments are conducted on two representative datasets: the large-scale synthetic dataset RESIDE and the small-scale real-world datasets I-HAZY and O-HAZY. The experimental results show that the proposed dehazing network is not only comparable to state-of-the-art methods for the RESIDE synthetic datasets, but also surpasses them by a very large margin for the I-HAZY and O-HAZY real-world dataset.

scholarly journals Evolutionary Social Poisson Factorizationfor Temporal Recommendation

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
ChunYan Yin ◽  
YongHeng Chen ◽  
Wanli Zuo
Keyword(s):  
Recommendation Systems ◽  
Predictive Performance ◽  
User Preferences ◽  
Social Model ◽  
User Preference ◽  
Latent Factors ◽  
Factorization Model ◽  
Real World Datasets ◽  
Item Attributes ◽  
The Impact

AbstractPreference-based recommendation systems analyze user-item interactions to reveal latent factors that explain our latent preferences for items and form personalized recommendations based on the behavior of others with similar tastes. Most of the works in the recommendation systems literature have been developed under the assumption that user preference is a static pattern, although user preferences and item attributes may be changed through time. To achieve this goal, we develop an Evolutionary Social Poisson Factorization (EPF$$\_$$ _ Social) model, a new Bayesian factorization model that can effectively model the smoothly drifting latent factors using Conjugate Gamma–Markov chains. Otherwise, EPF$$\_$$ _ Social can obtain the impact of friends on social network for user’ latent preferences. We studied our models with two large real-world datasets, and demonstrated that our model gives better predictive performance than state-of-the-art static factorization models.

scholarly journals Impact of external factors on quality financial reporting: The case of Vietnam

2016 ◽  
Vol 19 (1) ◽  
pp. 61-70
Author(s):  
Thuan Quoc Pham ◽  
Dao Xuan La
Keyword(s):  
Financial Reporting ◽  
External Factors ◽  
Internal Factors ◽  
Factors Affecting ◽  
Two Factors ◽  
Accounting Software ◽  
Major Factors ◽  
Pressure Factor ◽  
Independent Audit

The objective of this study is to identify the major factors that impact on the quality of financial reporting and to suggest model of the factors affecting on the quality of financial reporting for Vietnamese enterprises . There are two factors group affecting on the quality of financial reporting: internal factors and external factors, this study is limited to the scope of the external factors. By using case studies, analytical results indicate that Tax Pressure factor has the greatest impact on the quality of financial reporting, the remaning factors include: Listed Securities, Accounting Software and Independent Audit. Two factors, Politics and Regulatory Environment have limited impact on the quality financial reporting.

Sign in / Sign up

Export Citation Format

Share Document