Text Classification has received significant attention in recent years because of the proliferation of digital documents and is widely used in various applications such as filtering and recommendation. Consequently, many approaches, including those based on statistical theory, machine learning, and classifier performance improvement, have been proposed for improving text classification performance. Among these approaches, centroid-based classifier, multinomial naïve bayesian (MNB), support vector machines (SVM), convolutional neural network (CNN) are commonly used. In this paper, we introduce a cosine similarity-based methodology for improving performance. The methodology combines cosine similarity (between a test document and fixed categories) with conventional classifiers such as MNB, SVM, and CNN to improve the accuracy of the classifiers, and then we call the conventional classifiers with cosine similarity as enhanced classifiers. We applied the enhanced classifiers to famous datasets – 20NG, R8, R52, Cade12, and WebKB – and evaluated the performance of the enhanced classifiers in terms of the confusion matrix’s accuracy; we obtained outstanding results in that the enhanced classifiers show significant increases in accuracy. Moreover, through experiments, we identified which of two considered knowledge representation techniques (word count and term frequency-inverse document frequency (TFIDF)) is more suitable in terms of classifier performance.
In semiconductor manufacturing, not only maintaining a high yield, but also accurate yield prediction is highly important for improving productivity, customer satisfaction, and enhancing profitability. Despite of importance and merits, wafer yield prediction is a very challenging goal to be systematic with high quality and accuracy. In this paper, we propose a method for wafer edge yield prediction using a combined long short-term memory (LSTM) and feed-forward neural network (FFNN) model. Unlike previous research, we focus on the edge yield because of the higher yield loss at the wafer edge. The combined LSTM-FFNN model uses the dataset divided by two types according to its characteristics. Time-series data are used in the case of LSTM, and non-time-series data are fed into FFNN. When we prepared time-series data consisting of data relating to the equipment and chambers, no single data of different chamber overlaps with each other to regard them as independent entities. The proposed model outperforms the other comparing models in terms of all evaluation metrics. Coefficient of determination of the proposed combined LSTM-FFNN model is 34.14%, and this is almost 13% higher than the other comparing models on the average.
An effective supply chain management system is indispensable for an enterprise with a supply chain network in several aspects. Especially, organized control over the production and transportation of its products is a key success factor for the enterprise to stay active without damaging its reputation. This case is also highly relevant to garment industries. In this study, an extensive Deep Reinforcement Learning study for apparel supply chain optimization is proposed and undertaken, with focus given to Soft Actor-Critic. Six models are experimented with in this study and are compared with respect to the sell-through rate, service level, and inventory-to-sales ratio. Soft Actor-Critic outperformed several other state-of-the-art Actor Critic models in managing inventories and fulfilling demands. Furthermore, explicit indicators are calculated to assess the performance of the models in the experiment. Soft Actor-Critic achieved a better balance between service level and sell-through rate by ensuring higher availability of the stocks to sell without overstocking. From numerical experiments, it has been shown that S-policy, Trust Region Policy Optimization, and Twin Delayed Deep Deterministic Policy Gradient have a good balance between service level and sell-through rate. Additionally, Soft Actor-Critic achieved a 7%, 41.6%, and 42.8% lower inventory sales ratio than the S-policy, Twin Delayed Deep Deterministic Policy Gradient, and Trust Region Policy Optimization models, indicating its superior ability in making the inventory stocks available to make sales and profit from them.
8 January 2020 in Journal of Intelligent Information Systems
Text summarization is an important task in natural language processing (NLP). Neural summary models summarize information by understanding and rewriting documents through the encoder-decoder structure. Recent studies have sought to overcome the bias that cross-entropy-based learning methods can have through reinforcement learning (RL)-based learning methods or the problem of failing to learn optimized for metrics. However, the ROUGE metric with only $n$ -gram matching is not a perfect solution. The purpose of this study is to improve the quality of the summary statement by proposing a reward function used in text summarization based on RL. We propose ROUGE-SIM and ROUGE-WMD, modified functions of the ROUGE function. ROUGE-SIM enables meaningfully similar words, in contrast to ROUGE-L. ROUGE-WMD is a function adding semantic similarity to ROUGE-L. The semantic similarity between articles and summary text was computed using Word Mover’s Distance (WMD) methodology. Our model with two proposed reward functions demonstrated superior performance on ROUGE-1, ROUGE-2, and ROUGE_L than on ROUGE-L as a reward function. Our two models, ROUGE-SIM and ROUGE-WMD, scored 0.418 and 0.406 for ROUGE-L, respectively, for the Gigaword dataset. The two reward functions outperformed ROUGE-L even in the abstractiveness and grammatical aspects.