Recommender Systems using SVD with Social Network Information

Kim, Min-Gun; Kim, Kyoung-jae

doi:10.13088/jiis.2016.22.4.001

Cited by 4 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Online social platforms enable remote communication between digital creators, consumers, advertisers, diverse entities, and the platform itself (Kim, 2022). The platform facilitates the generation of value for users by providing virtual space, user-friendly tools, and various ways.…”

Section: Theoretical Framework For Digital Creator Ecosystemmentioning

confidence: 99%

Exploring online social platform affordances for digital creators: a multi-method approach using qualitative and configurational analysis

Lee,

Chang,

Lee

et al. 2024

IMDS

View full text Add to dashboard Cite

PurposeThis study explores the key platform affordances that online social platform providers need to offer digital creators to strengthen the creator ecosystem, one of the leading accelerators for platform growth. Specifically, it aims to investigate how these affordances make the dynamic combinations for high platform quality across diverse platform types and demographic characteristics of digital creators.Design/methodology/approachThis study adopts a multi-method approach. Drawing upon the affordance theory, Study 1 aims to identify the key affordances of online social platforms based on relevant literature and the qualitative interview data collected from 22 digital creators, thereby constructing a conceptual framework of key platform affordances for digital creators. Building on the findings of Study 1, Study 2 explores the dynamic combinations of these platform affordances that contribute to platform quality using a configurational approach. Data from online surveys of 185 digital creators were analyzed using fuzzy set qualitative comparative analysis (fsQCA).FindingsThe results of Study 1 identified key online social platform affordances for digital creators, including Storytelling, Socialization, Design, Development, Promotion, and Protection affordance. Study 2 showed that the combinations of these platform affordances for digital creators are diverse according to the types of platforms, creators’ gender, and their professionality.Research limitations/implicationsLike many studies, this research also has several limitations. One limitation of the research is the potential constraint of the extent of how well the data samples represent the group of creators who are actively producing digital content. Despite the addition of screening questions and meticulous data filtering, it is possible that we did not secure sufficient data from creators who are actively engaged in creative activities. In future research, it is worth contemplating the acquisition of data from actual groups of creators, such as creator communities. Future researchers anticipate obtaining more in-depth and accurate data by directly involving and collaborating with creators.Practical implicationsThis study highlights the need for online social platforms to enhance features for storytelling, socializing, design, development, promotion, and protection, fostering a robust digital creator ecosystem. It emphasizes clear communication of these affordances, ensuring creators can effectively utilize them. Importantly, platforms should adapt these features to accommodate diverse creator profiles, considering differences in gender and expertise levels, especially in emerging spaces like the Metaverse. This approach ensures an equitable and enriching experience for all users and creators, underlining the importance of dynamic interaction and inclusivity in platform development and creator support strategies.Social implicationsThis study underscores the social implications of evolving digital creator ecosystems on online platforms. Identifying six key affordances essential for digital creators highlights the need for platforms to enhance storytelling, socializing, design, development, promotion, and product protection. Crucially, it emphasizes inclusivity, urging platforms to consider diverse creator profiles, including gender and expertise differences, particularly in transitioning from traditional social media to the Metaverse. This approach nurtures a more robust creator ecosystem and fosters an equitable and enriching experience for all users. It signals a shift towards more dynamic, adaptive online environments catering to diverse creators and audiences.Originality/valueFor academics, this study builds the conceptual framework of online social platform affordances for digital creators. Using the configurational approach, this study identified various interdependent relationships among the affordances, which are nuanced by specific contexts, and suggested novel insights for future studies. For practices, the findings specified by creators and platform types are expected to guide platform providers in developing strategies to support digital creators and contribute to platform growth.

show abstract

Section: Theoretical Framework For Digital Creator Ecosystemmentioning

confidence: 99%

Exploring online social platform affordances for digital creators: a multi-method approach using qualitative and configurational analysis

Lee,

Chang,

Lee

et al. 2024

IMDS

View full text Add to dashboard Cite

show abstract

“…A traditional recommendation system involves users, items, and transactions between users and items. A user-item rating matrix is used as a knowledge source of the RS [9]. Some RS methods employ users' ratings as the only source of information and they do not consider any additional information.…”

Section: Related Workmentioning

confidence: 99%

“…SVD can alleviate the sparsity problem by constructing a low-dimensional matrix. SVD was used for collaborative filtering in the Netflix competition to realize the objectives of recommendation algorithms [9]. Sarwar et al [39] proposed a method in which the score of the prediction is computed after the dimension of movie data has been decreased via SVD.…”

Section: Singular Value Decomposition (Svd)mentioning

confidence: 99%

“…Another issue is the cold-start problem, which refers to the problem of making predictions for a new customer or item: The RS cannot recommend items to a new user because there is no history regarding his/her tastes. Likewise, the RS lacks information regarding the rate or score of a cold-start item [9]. Cold-start is divided into two types: incomplete cold start (ICS) and complete cold start (CCS).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Exploiting implicit social relationships via dimension reduction to improve recommendation system performance

Al-Sabaawi¹,

Karacan²,

Yenice³

2020

PLoS ONE

View full text Add to dashboard Cite

The development of Web 2.0 and the rapid growth of available data have led to the development of systems, such as recommendation systems (RSs), that can handle the information overload. However, RS performance is severely limited by sparsity and cold-start problems. Thus, this paper aims to alleviate these problems. To realize this objective, a new model is proposed by integrating three sources of information: a user-item matrix, explicit and implicit relationships. The core strategy of this study is to use the multi-step resource allocation (MSRA) method to identify hidden relations in social information. First, explicit social information is used to compute the similarity between each pair of users. Second, for each nonfriend pair of users, the MSRA method is applied to determine the probability of their relation. If the probability exceeds a threshold, a new relationship will be established. Then, all sources are incorporated into the Singular Value Decomposition (SVD) method to compute the missing prediction values. Furthermore, the stochastic gradient descent technique is applied to optimize the training process. Additionally, two real datasets, namely, Last.Fm and Ciao, are utilized to evaluate the proposed method. In terms of accuracy, the experiment results demonstrate that the proposed method outperforms eight state-of-the-art approaches: Heats, PMF, SVD, SR, EISR-JC, EISR-CN, EISR-PA and EISR-RAI.

show abstract

“…For example, a recommendation method based on similarity analysis is proposed in [17]. In [18], a user interest point recommendation method is proposed based on useritem score matrix, which is constructed and calculated by users and their interest points. Reference [19] takes the user scores as recommendation basis and generate a list of recommended items for each user to address the cold start problem.…”

Section: Introductionmentioning

confidence: 99%

Interest Points Analysis for Internet Forum Based on Long-Short Windows Similarity

Ju¹,

Lu²,

Luo³

et al. 2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

For Internet forum Points of Interest (PoI), existing analysis methods are usually lack of usability analysis under different conditions and ignore the long-term variation, which lead to blindness in method selection. To address this problem, this paper proposed a PoI variation prediction framework based on similarity analysis between long and short windows. Based on the framework, this paper presented 5 PoI analysis algorithms which can be categorized into 2 types, i.e., the traditional sequence analysis methods such as autoregressive integrated moving average model (ARIMA), support vector regressor (SVR), and the deep learning methods such as convolutional neural network (CNN), long-short term memory network (LSTM), Transformer (TRM). Specifically, this paper firstly divides observed data into long and short windows, and extracts key words as PoI of each window. Then, the PoI similarities between long and short windows are calculated for training and prediction. Finally, series of experiments is conducted based on real Internet forum datasets. The results show that, all the 5 algorithms could predict PoI variations well, which indicate effectiveness of the proposed framework. When the length of long window is small, traditional methods perform better, and SVR is the best. On the contrary, the deep learning methods show superiority, and LSTM performs best. The results could provide beneficial references for PoI variation analysis and prediction algorithms selection under different parameter configurations.

show abstract

Recommender Systems using SVD with Social Network Information

Cited by 4 publications

References 12 publications

Exploring online social platform affordances for digital creators: a multi-method approach using qualitative and configurational analysis

Exploring online social platform affordances for digital creators: a multi-method approach using qualitative and configurational analysis

Exploiting implicit social relationships via dimension reduction to improve recommendation system performance

Interest Points Analysis for Internet Forum Based on Long-Short Windows Similarity

Contact Info

Product

Resources

About