Digital Library[ Search Result ]
Effect of Denoising Autoencoder in the view of Item Popularity Bias
Jinhong Kim, Jae-woong Lee, Jongwuk Lee
http://doi.org/10.5626/JOK.2021.48.5.575
Denoising autoencoder (DAE) is commonly used in recent recommendation systems. It is a type of Autoencoder that trains by giving noise to the input and has shown improved performance compared to autoencoder. In this paper, we analyze the effect of noise in terms of item popularity to interpret the training of DAE. For analysis, we design the experiment in the following two ways. First, we observe the changes of the learned item vector’s L2-norm by giving noise to the autoencoder. Second, by giving noise only to presampled items by popularity, we anlayze whether the improved performance of the DAE is related to item popularity. Results of the experiment showed that the variance of the item vector norm caused by popularity was reduced by noise, and that the accuracy increased when noise was given to the popular items.
Performance Comparison and Analysis Between Neural and Non-neural Autoencoder-based Recommender Systems
http://doi.org/10.5626/JOK.2020.47.11.1078
While deep neural networks have been bringing advances in many domains, recent studies have shown that the performance gain from deep neural networks is not as extensive as reported, compared to the higher computational complexity they require. This phenomenon is caused by the lack of shared experimental settings and strict analysis of proposed methods. In this paper, 1) we build experimental settings for fair comparison between the different recommenders, 2) provide empirical studies on the performance of the autoencoder-based recommender, which is one of the main families in the literature, and 3) analyze the performance of a model according to user and item popularity. With extensive experiments, we found that there was no consistent improvement between the neural and the non-neural models in every dataset and there is no evidence that the non-neural models have been improving over time. Also, the non-neural models achieved better performance on popular item accuracy, while the neural models relatively perform better on less popular items.
A CNN-based Column Prediction Model for Generating SQL Queries using Natural Language
Yoonki Jeong, Dongmin Kim, Jongwuk Lee
http://doi.org/10.5626/JOK.2019.46.2.202
To retrieve massive data using relational database management system (RDBMS), it is important to understanding of table schemas and SQL grammar. To address this issue, many studies have recently been carried out to generate an SQL query from a natural language question. However, the existing works suffer mostly from predicting columns at where clause and the accuracy is greatly reduced when there are multiple columns to be predicted. In this paper, we propose a convolutional neural network model with column attention mechanism that effectively extracts the latent representation of input question which helps column prediction of the model. The experiment shows that our model outperforms the accuracy of the existing model (SQLNet) by 6%.
Correct Linear Skyline Algorithm in High-Dimensional Space
http://doi.org/10.5626/JOK.2018.45.10.1089
Skyline query is a preference query that finds a candidate set for user preferences, employing the dominance property. It can be effectively used for decision problems that have multiple data attributes. However, a problem arises whereby the skyline becomes too large as the number of attributes in the data increases. To solve this problem, in this paper, we propose a new algorithm for a linear skyline query that restricts a user’s preference function by a linear function. In the previous work, a method was proposed to obtain a linear skyline by adding the same number of virtual points to the data as the number of attributes. However, it has been observed that this previous method does not guarantee the correctness of the linear skyline. We revised this method by adding virtual points in order to find the correct linear skyline. We prove that the proposed algorithm finds the correct linear skyline, and we empirically evaluate the correctness of the proposed algorithm.
An Effective Preference Model to Improve Top-N Recommendation
http://doi.org/10.5626/JOK.2017.44.6.621
Collaborative filtering is a technique that effectively recommends unrated items for users. Collaborative filtering is based on the similarity of the items evaluated by users. The existing top-N recommendation methods are based on pair-wise and list-wise preference models. However, these methods do not effectively represent the relative preference of items that are evaluated by users, and can not reflect the importance of each item. In this paper, we propose a new method to represent user"s latent preference by combining an existing preference model and the notion of inverse user frequency. The proposed method improves the accuracy of existing methods by up to two times.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr