A L1 Regularized Logistic Regression Model for Highdimensional Questionnaire Data Analysis

The LI regularization method, or Lasso, is a technique for feature selection in high-dimensional statistical analysis. This method compresses the coefficients of the model by using the absolute value of the coefficient function as a penalty term. By adding L1 regularization to log-likelihood function of Logistic model, variable screening method based on the logistic regression model can be realized. The process of variable selection via Lasso is illustrated in Figure 1. The purpose of the experiment is to figure out the important factors that influence interviewees' subjective well-being using L1 regularized logistic regression. Experiments have been performed on CGSS 2017 data. Important features have been successfully selected by using the L1 regularization method.

ILRC: a hybrid biomarker discovery algorithm based on improved L1 regularization and clustering in microarray data

Background Finding significant genes or proteins from gene chip data for disease diagnosis and drug development is an important task. However, the challenge comes from the curse of the data dimension. It is of great significance to use machine learning methods to find important features from the data and build an accurate classification model. Results The proposed method has proved superior to the published advanced hybrid feature selection method and traditional feature selection method on different public microarray data sets. In addition, the biomarkers selected using our method show a match to those provided by the cooperative hospital in a set of clinical cleft lip and palate data. Method In this paper, a feature selection algorithm ILRC based on clustering and improved L1 regularization is proposed. The features are firstly clustered, and the redundant features in the sub-clusters are deleted. Then all the remaining features are iteratively evaluated using ILR. The final result is given according to the cumulative weight reordering. Conclusion The proposed method can effectively remove redundant features. The algorithm’s output has high stability and classification accuracy, which can potentially select potential biomarkers.

Evaluation of Weight Decay Regularization Techniques for Stock Price Prediction using Gated Recurrent Unit Network

Stock price forecasting in the field of interest for many stock investors to earn more profit from stock trading. Nowadays, machine learning researchers are also involved in this research field so that fast, accurate and automatic stock price forecasting can be achieved. This research paper evaluated GRU network’s performance with weight decay reg-ularization techniques for predicting price of stocks listed NEPSE. Three weight decay regularization technique analyzed in this research work were (1) L1 regularization (2) L2 regularization and (3) L1_L2 regularization. In this research work, six randomly selected stocks from NEPSE were experimented. From the experimental results, we observed that L2 regularization could outperform L1 and L1_L2 reg-ularization techniques for all six stocks. The average MSE obtained with L2 regularization was 4.12% to 33.52% lower than the average MSE obtained with L1 regularization, and it was 10.92% to 37.1% lower than the average MSE obtained with L1_L2 regularization. Thus, we concluded that the L2 regularization is best choice among weight regularization for stock price prediction.

Development of an Abbreviated Adult Reading History Questionnaire (ARHQ-Brief) Using a Machine Learning Approach

Several crucial reasons exist to identify whether an adult has had reading disorder (RD) and to predict a child’s likelihood of developing RD. The Adult Reading History Questionnaire (ARHQ) is among the most commonly used self-reported questionnaires. High ARHQ scores indicate an increased likelihood that an adult had RD as a child, and that their children may develop RD. This study focused on whether a subset of ARHQ items (ARHQ-brief) could be equally effective in assessing adults’ reading history as the full ARHQ. We used a machine learning approach, lasso (known as L1 regularization), and identified 6 of 23 items that resulted in the ARHQ-brief. Data from 97 adults and 47 children were included. With the ARHQ-brief, we report a threshold of 0.323 as suitable to identify past likelihood of RD in adults with a sensitivity of 72.4% and a specificity of 81.5%. Comparison of predictive performances between ARHQ-brief and the full ARHQ showed that ARHQ-brief explained an additional 10%–35.2% of the variance in adult and child reading. Furthermore, we validated ARHQ-brief’s superior ability to predict reading ability using an independent sample of 28 children. We close by discussing limitations and future directions.

Deblurring Turbulent Images via Maximizing L1 Regularization

Atmospheric turbulence significantly degrades image quality. A blind image deblurring algorithm is needed, and a favorable image prior is the key to solving this problem. However, the general sparse priors support blurry images instead of explicit images, so the details of the restored images are lost. The recently developed priors are non-convex, resulting in complex and heuristic optimization. To handle these problems, we first propose a convex image prior; namely, maximizing L1 regularization (ML1). Benefiting from the symmetrybetween ML1 and L1 regularization, the ML1 supports clear images and preserves the image edges better. Then, a novel soft suppression strategy is designed for the deblurring algorithm to inhibit artifacts. A coarse-to-fine scheme and a non-blind algorithm are also constructed. For qualitative comparison, a turbulent blur dataset is built. Experiments on this dataset and real images demonstrate that the proposed method is superior to other state-of-the-art methods in blindly recovering turbulent images.

ILRC : A Hybrid Biomarker Discovering Algorithm Based on Improved L1 Regularization and Clustering in Microarray Data

Background: Finding significant genes or proteins from gene chip data for disease diagnosis and drug development is an important task, and the challenge comes from the curse of the data dimension. It is of great significance to use machine learning methods to find important features from the data and build an accurate classification model. Results: The proposed Mehtod has proved superior to the published advanced hybrid feature selection method and traditional feature selection method on different public microarray data sets. In addition, the results on the cleft lip and palate data set with known biomarkers provided by the cooperative hospital show that compared with other methods, our method can preferentially select these biomarkers. Method: In this paper, a feature selection algorithm ILRC based on clustering and improved L1 regularization is proposed. In this method, the features are first clustered, and the redundant features in the sub-clusters are deleted. Then all the remaining features are iteratively evaluated using ILR, and the final result is output according to the cumulative weight reordering. Conclusion: The proposed method can effectively remove redundant features. The algorithm’s output has high stability and classification accuracy and can potentially select potential biomarkers.