Search results

In this “Info-plosion” era, recommendation systems (or recommenders) play a significant role in finding interesting items in the surge of online digital activity and e-commerce. The purpose of this paper is to model users' preference evolution to recommend potential items which users may be interested in.

A novel recommendation system, namely evolution-learning recommendation (ELR), is developed to precisely predict user interest for making recommendations. Differing from prior related methods, the authors integrate the matrix factorization (MF) and recurrent neural network (RNN) to effectively describe the variation of user preferences over time.

A novel cumulative factorization technique is proposed to efficiently decompose a rating matrix for discovering latent user preferences. Compared to traditional MF-based methods, the cumulative MF could reduce the utilization of computation resources. Furthermore, the authors depict the significance of long- and short-term effects in the memory cell of RNN for evolution patterns. With the context awareness, a learning model, V-LSTM, is developed to dynamically capture the evolution pattern of user interests. By using a well-trained learning model, the authors predict future user preferences and recommend related items.

Based on the relations among users and items for recommendation, the authors introduce a novel concept, virtual communication, to effectively learn and estimate the correlation among users and items. By incorporating the discovered latent features of users and items in an evolved manner, the proposed ELR model could promote “right” things to “right” users at the “right” time. In addition, several extensive experiments are performed on real datasets and are discussed. Empirical results show that ELR significantly outperforms the prior recommendation models. The proposed ELR exhibits great generalization and robustness in real datasets, including e-commerce, industrial retail and streaming service, with all discussed metrics.

The purpose of this paper is to provide a cross-platform recommendation system that recommends the most suitable public Instagram accounts to Facebook users.

We collect data from both Facebook and Instagram and then propose a similarity matching mechanism for recommending the most appropriate Instagram accounts to Facebook users. By removing the data disparity between the two heterogeneous platforms and integrating them, the system is able to make more accurate recommendations.

The results show that the method proposed in this paper can recommend suitable public Instagram accounts to Facebook users with very high accuracy.

To the best of the authors’ knowledge, this is the first study to propose a recommender system to recommend Instagram public accounts to Facebook users. Second, our proposed method can integrate heterogeneous data from two different platforms to generate collaborative recommendations. Furthermore, our cross-platform system reveals an innovative concept of how multiple platforms can promote their respective platforms in a unified, cooperative and collaborative manner.

A necessary preprocessing of document classification is to label some documents so that a classifier can be built based on which the remaining documents can be classified. Because each document differs in length and complexity, the cost of labeling each document is different. The purpose of this paper is to consider how to select a subset of documents for labeling with a limited budget so that the total cost of the spending does not exceed the budget limit, while at the same time building a classifier with the best classification results.

In this paper, a framework is proposed to select the instances for labeling that integrate two clustering algorithms and two centroid selection methods. From the selected and labeled instances, five different classifiers were constructed with good classification accuracy to prove the superiority of the selected instances.

Experimental results show that this method can establish a training data set containing the most suitable data under the premise of considering the cost constraints. The data set considers both “data representativeness” and “data selection cost,” so that the training data labeled by experts can effectively establish a classifier with high accuracy.

No previous research has considered how to establish a training set with a cost limit when each document has a distinct labeling cost. This paper is the first attempt to resolve this issue.

As researchers are writing a draft paper with incomplete structure or text, one of burdensome tasks is to deliberate about which references should be cited for one sentence or paragraph of this draft. In view of the rapid increase in the number of research papers, researchers desire to figure out a better way to do citation recommendations in developing their draft papers. The purpose of this paper is to propose citation recommendation algorithms that enable the acquisition of relevant citations for research papers that are still at the drafting stage. This study attempts to help researchers to select appropriate references among the vast amount of available papers and make draft papers complete in reference citation.

This study adopts a model for recommending citations for incomplete drafts. Four algorithms are proposed in this study. The first and second algorithms are unsupervised models, applying term frequency-inverse document frequency and WordNet technologies, respectively. The third and fourth algorithms are based on the second algorithm to integrate different weight adjustment strategies to improve performance.

The proposed recommendation method adopts three techniques, including using WordNet to transform vector and setting adjustment weights according to structural factors and the information completeness degree, to generate citation recommendation for incomplete drafts. The experiments show that all these three techniques can significantly improve the recommendation accuracy.

None of the methods employed in previous studies can recommend articles as references for incomplete drafts. This paper addresses the situation that a draft paper can be incomplete either in structure or text or both. Recommended references, however, can be still generated and inserted into any desired sentence of the draft paper.

The purpose of this paper is to determine the best approach to customer segmentation and to extrapolate associated rules for this based on recency, frequency and monetary (RFM) considerations as well as demographic factors. In this study, the impacts of RFM and demographic attributes have been challenged in order to enrich factors that lend comprehension to customer segmentation. Different types of scenario were designed, performed and evaluated meticulously under uniform test conditions. The data for this study were extracted from the database of a global pizza restaurant chain in Turkey. This paper summarizes the findings of the study and also provides evidence of its empirical implications to improve the performance of customer segmentation as well as achieving extracted rule perfection via effective model factors and variations. Accordingly, marketing and service processes will work more effectively and efficiently for customers and society. The implication of this study is that it explains a clear concept for interaction between producers and consumers.

Customer relationship management, which aims to manage record and evaluate customer interactions, is generally regarded as a vital tool for companies that wish to be successful in the rapidly changing global market. The prediction of customer behaviors is a strategically important and difficult issue because of the high variance and wide range of customer orders and preferences. So to have an effective tool for extracting rules based on customer purchasing behavior, considering tangible and intangible criteria is highly important. To overcome the challenges imposed by the multifaceted nature of this problem, the authors utilized artificial intelligence methods, including k-means clustering, Apriori association rule mining (ARM) and neural networks. The main idea was that customer clusters are better enhanced when segmentation processes are based on RFM analysis accompanied by demographic data. Weighted RFM (WRFM) and unweighted RFM values/scores were applied with and without demographic factors and utilized to compose different types and numbers of clusters. The Apriori algorithm was used to extract rules of association. The performance analyses of scenarios have been conducted based on these extracted rules. The number of rules, elapsed time and prediction accuracy were used to evaluate the different scenarios. The results of evaluations were compared with the outputs of another available technique.

The results showed that having an appropriate segmentation approach is vital if there are to be strong association rules. Also, it has been determined from the results that the weights of RFM attributes affect rule association performance positively. Moreover, to capture more accurate customer segments, a combination of RFM and demographic attributes is recommended for clustering. The results’ analyses indicate the undeniable importance of demographic data merged with WRFM. Above all, this challenge introduced the best possible sequence of factors for an analysis of clustering and ARM based on RFM and demographic data.

The work compared k-means and Kohonen clustering methods in its segmentation phase to prove the superiority of adopted segmentation techniques. In addition, this study indicated that customer segments containing WRFM scores and demographic data in the same clusters brought about stronger and more accurate association rules for the understanding of customer behavior. These so-called achievements were compared with the results of classical approaches in order to support the credibility of the proposed methodology. Based on previous works, classical methods for customer segmentation have overlooked any combination of demographic data with WRFM during clustering before proceeding to their rule extraction stages.

Study on knowledge-based systems for scientific publications is growing very broadly. However, most of these studies do not explicitly discuss the knowledge management (KM) component as knowledge management system (KMS) implementation. This background causes academic institutions to face challenges in developing KMS to support scholarly publication cycle (SPC). Therefore, this study aims to develop a new KMS conceptual model, Identify critical components and provide research gap opportunities for future KM studies on SPC.

This study used a systematic literature review (SLR) method with the procedure from Kitchenham et al. Then, the SLR results are compiled into a conceptual model design based on a framework on KM foundations and KM solutions. Finally, the model design was validated through interviews with related field experts.

The KMS for SPC focuses on the discovery, sharing and application of knowledge. The majority of KMS use recommendation systems technology with content-based filtering and collaborative filtering personalization approaches. The characteristics data used in KMS for SPC are structured and unstructured. Metadata and article abstracts are considered sufficiently representative of the entire article content to be used as a search tool and can provide recommendations. The KMS model for SPC has layers of KM infrastructure, processes, systems, strategies, outputs and outcomes.

This study has limitations in discussing tacit knowledge. In contrast, tacit knowledge for SPC is essential for scientific publication performance. The tacit knowledge includes experience in searching, writing, submitting, publishing and disseminating scientific publications. Tacit knowledge plays a vital role in the development of knowledge sharing system (KSS) and KCS. Therefore, KSS and KCS for SPC are still very challenging to be researched in the future. KMS opportunities that might be developed further are lessons learned databases and interactive forums that capture tacit knowledge about SPC. Future work potential could identify other types of KMS in academia and focus more on SPC.

This study proposes a novel comprehensive KMS model to support scientific publication performance. This model has a critical path as a KMS implementation solution for SPC. This model proposes and recommends appropriate components for SPC requirements (KM processes, technology, methods/techniques and data). This study also proposes novel research gaps as KMS research opportunities for SPC in the future.