scholarly journals An efficient stock market prediction model using hybrid feature reduction method based on variational autoencoders and recursive feature elimination

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Hakan Gunduz
Keyword(s):  
Reduction Method ◽  
Feature Reduction ◽  
The Other ◽  
Recursive Feature Elimination ◽  
Accuracy Rate ◽  
Stock Performance ◽  
Attention Model ◽  
Feature Sets ◽  
Accuracy Rates ◽  
F Measure

AbstractIn this study, the hourly directions of eight banking stocks in Borsa Istanbul were predicted using linear-based, deep-learning (LSTM) and ensemble learning (LightGBM) models. These models were trained with four different feature sets and their performances were evaluated in terms of accuracy and F-measure metrics. While the first experiments directly used the own stock features as the model inputs, the second experiments utilized reduced stock features through Variational AutoEncoders (VAE). In the last experiments, in order to grasp the effects of the other banking stocks on individual stock performance, the features belonging to other stocks were also given as inputs to our models. While combining other stock features was done for both own (named as allstock_own) and VAE-reduced (named as allstock_VAE) stock features, the expanded dimensions of the feature sets were reduced by Recursive Feature Elimination. As the highest success rate increased up to 0.685 with allstock_own and LSTM with attention model, the combination of allstock_VAE and LSTM with the attention model obtained an accuracy rate of 0.675. Although the classification results achieved with both feature types was close, allstock_VAE achieved these results using nearly 16.67% less features compared to allstock_own. When all experimental results were examined, it was found out that the models trained with allstock_own and allstock_VAE achieved higher accuracy rates than those using individual stock features. It was also concluded that the results obtained with the VAE-reduced stock features were similar to those obtained by own stock features.

scholarly journals Prediction of Head Movement in 360-Degree Videos Using Attention Model

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3678
Author(s):  
Dongwon Lee ◽  
Minji Choi ◽  
Joohyun Lee
Keyword(s):  
Machine Learning ◽  
Short Term Memory ◽  
Moving Average ◽  
The Other ◽  
Learning Models ◽  
Short Term ◽  
Term Memory ◽  
Attention Model ◽  
Long Short Term Memory ◽  
Machine Learning Models

In this paper, we propose a prediction algorithm, the combination of Long Short-Term Memory (LSTM) and attention model, based on machine learning models to predict the vision coordinates when watching 360-degree videos in a Virtual Reality (VR) or Augmented Reality (AR) system. Predicting the vision coordinates while video streaming is important when the network condition is degraded. However, the traditional prediction models such as Moving Average (MA) and Autoregression Moving Average (ARMA) are linear so they cannot consider the nonlinear relationship. Therefore, machine learning models based on deep learning are recently used for nonlinear predictions. We use the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) neural network methods, originated in Recurrent Neural Networks (RNN), and predict the head position in the 360-degree videos. Therefore, we adopt the attention model to LSTM to make more accurate results. We also compare the performance of the proposed model with the other machine learning models such as Multi-Layer Perceptron (MLP) and RNN using the root mean squared error (RMSE) of predicted and real coordinates. We demonstrate that our model can predict the vision coordinates more accurately than the other models in various videos.

scholarly journals Augmented Data Selector to Initiate Text-Based CAPTCHA Attack

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Aolin Che ◽  
Yalin Liu ◽  
Hong Xiao ◽  
Hao Wang ◽  
Ke Zhang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Training Data ◽  
Accuracy Rate ◽  
Data Set ◽  
Research Directions ◽  
Training Models ◽  
The Past ◽  
Future Work ◽  
Design Cost ◽  
Accuracy Rates

In the past decades, due to the low design cost and easy maintenance, text-based CAPTCHAs have been extensively used in constructing security mechanisms for user authentications. With the recent advances in machine/deep learning in recognizing CAPTCHA images, growing attack methods are presented to break text-based CAPTCHAs. These machine learning/deep learning-based attacks often rely on training models on massive volumes of training data. The poorly constructed CAPTCHA data also leads to low accuracy of attacks. To investigate this issue, we propose a simple, generic, and effective preprocessing approach to filter and enhance the original CAPTCHA data set so as to improve the accuracy of the previous attack methods. In particular, the proposed preprocessing approach consists of a data selector and a data augmentor. The data selector can automatically filter out a training data set with training significance. Meanwhile, the data augmentor uses four different image noises to generate different CAPTCHA images. The well-constructed CAPTCHA data set can better train deep learning models to further improve the accuracy rate. Extensive experiments demonstrate that the accuracy rates of five commonly used attack methods after combining our preprocessing approach are 2.62% to 8.31% higher than those without preprocessing approach. Moreover, we also discuss potential research directions for future work.

scholarly journals Treynor Ratio to Measure Islamic Stock Performance in Indonesia

2019 ◽  
Vol 8 (1) ◽  
pp. 1-13
Author(s):  
Mohammad Farhan Qudratullah
Keyword(s):  
Interest Rates ◽  
Stock Returns ◽  
Capital Market ◽  
The Other ◽  
Islamic Finance ◽  
Ratio Model ◽  
Stock Performance ◽  
Measurement Results ◽  
Domestic Products ◽  
Very High

Treynor Ratio merupakan model pioner inovatif ukuran kinerja saham yang dikemukakan Jack Treynor pada tahun 1965 yang terdiri atas 3 (tiga) komponen, yaitu return saham, return bebas risiko, dan beta saham. Banyak penelitian mendekati return bebas risiko dengan suku bunga termasuk saat mengukur kinerja saham syariah, sedangkan suku bunga dilarang dalam konsep keuangan islam. Tulisan ini membahas variabel alternatif untuk mendekati return bebas risiko selain dengan suku bunga (BI-Rate), yaitu dengan 4 (empat) pendekatan, yaitu: menghilangkan suku bunga, mengganti dengan zakat rate, mengganti dengan inflasi, dan mengganti dengan gross domestic produc (GDP) pada model Treynor Ratio yang diimplementasikan pada pasar modal syariah di Indonesia periode Januari 2011-Juli 2018. Hasil yang diperoleh adalah terdapat kesesuaian yang sangat tinggi hasil pengukuran model Treynor Ratio dengan suku bunga dengan keempat model lainnya. Namun, model-model tersebut tidak menjamin bahwa saham yang memilki kinerja terbaik pada saat ini akan memilki kinerja terbaik dimasa yang akan datang atau sebaliknya. Dilihat dari kedekatan hasil pengukuran kinerjanya, kelima model Treynor Ratio tersebut dapat dikelompokan jadi 2 (dua), yaitu model dengan suku bunga, model dengan inflasi, dan model dengan GDP sebagai kelompok pertama, sedangkan model tanpa suku bunga dan model dengan zakat-rate sebagai kelompok kedua. [Treynor Ratio is an innovative pioneer model the size of stock performance proposed by Jack Treynor in 1965 which consists of 3 (three) components, namely stock returns, risk free returns, and stock beta. Many studies approach risk-free returns with interest rates, including when measuring the performance of Islamic stocks, while interest rates are prohibited in the concept of Islamic finance. This paper discusses alternative variables to approach risk-free returns other than interest rates (BI-Rate), namely with 4 (four) approaches, namely: eliminating interest rates, changing zakat rates, changing inflation, and substituting gross domestic products (GDP) in the Treynor Ratio model that is implemented in the Islamic capital market in Indonesia for the period January 2011 - July 2018. The results obtained are very high conformity in the measurement results of the Treynor Ratio model with interest rates with the other four models. However, these models do not guarantee that stocks that have the best performance at this time will have the best performance in the future or vice versa. Judging from the closeness of the results of performance measurement, the five Treynor Ratio models can be grouped into 2 (two), namely models with interest rates, models with inflation, and models with GDP as the first group, while models without interest rates and models with zakat-rate as second group.]

scholarly journals Reduction method of residual voltage by connecting lead length of surge protection device

2018 ◽  
Vol 7 (3.3) ◽  
pp. 384
Author(s):  
Young Dal Kim ◽  
Young Chan Kim ◽  
Yun Mi Jeong ◽  
Dae Dong Lee
Keyword(s):  
Reduction Method ◽  
The Other ◽  
Installation Space ◽  
Lead Wire ◽  
Protection Device ◽  
Design Technique ◽  
Wire Length ◽  
Residual Voltage ◽  
Protection Level ◽  
Surge Protection

Background/Objectives: In order to minimize the damage and malfunction of the equipment and system from various surges, we studied the method of reducing the residual voltage according to the lead wire length of the surge protector.Methods/Statistical analysis: In buildings, SPD installation space is insufficient or narrow, resulting in longer lead wire of SPD, and SPD protection performance is decreased due to increase of voltage protection level and residual voltage. In this study, the voltage protection level and the residual voltage of the conventional SPD model and the proposed SPD model are analyzed according to the change of the connecting conductor length from 0.5to 100m.Findings: In the case of the conventional SPD model, the protection level of the SPD is excellent by measuring the voltage protection level at 1,410V even if the lead wire length of the connecting conductor is changed to 10m, but when it exceeds 10m, the protection performance and the protection cooperation are reduced. On the other hand, in the case of the proposed SPD model, the voltage protection level was measured to be 50 V or less even if the lead wire length of the connecting conductor was changed to100 m. Therefore, it is considered that SPD protection performance and protection cooperation are excellent.Improvements/Applications: The design technique of SPD obtained through this study will help to select the optimal installation site and reduce the budget.  

RECOGNITION OF LEXICAL TONES FOR ISOLATED SYLLABLES AND DISYLLABLES IN MANDARIN SPEECH

1988 ◽  
Vol 02 (01) ◽  
pp. 49-69 ◽  
Author(s):  
WU-JI YANG ◽  
JYH-CHYANG LEE ◽  
YUEH-CHIN CHANG ◽  
HSIAO-CHUAN WANG
Keyword(s):  
Vector Quantization ◽  
Markov Models ◽  
Hidden Markov ◽  
Pitch Contour ◽  
Pitch Height ◽  
Accuracy Rate ◽  
Lexical Tones ◽  
Feature Vectors ◽  
Speaker Independent ◽  
Accuracy Rates

This study purposes a method for recognizing the lexical tones in Mandarin speech. The method is based on Vector Quantization (VQ) and Hidden Markov Models (HMM). The pitch periods are extracted to derive the feature vectors which represent pitch height and pitch contour slope. One HMM is trained by the feature vectors of monosyllables for each tone. Then the HMMs are used to recognize the tone of monosyllables and disyllables. For the monosyllables, the accuracy rate can be 93.75% for speaker-independent cases. For the disyllables, the accuracy rates are 93% for the first syllables and 90% for the second syllables. It shows that the tone of the second syllable may be affected by the preceding syllable. This degradation also reveals the fact of tone variation in Mandarin speech.

A proposed scheme for sentiment analysis

Kybernetes ◽  
2018 ◽  
Vol 47 (5) ◽  
pp. 957-984 ◽  
Author(s):  
Sajjad Tofighy ◽  
Seyed Mostafa Fakhrahmad
Keyword(s):  
Sentiment Analysis ◽  
Clustering Algorithm ◽  
Reduction Method ◽  
Feature Reduction ◽  
Sentiment Classification ◽  
Support Vector ◽  
Second Phase ◽  
Reduction Algorithm ◽  
Content Type ◽  
Statistical Knowledge

Purpose This paper aims to propose a statistical and context-aware feature reduction algorithm that improves sentiment classification accuracy. Classification of reviews with different granularities in two classes of reviews with negative and positive polarities is among the objectives of sentiment analysis. One of the major issues in sentiment analysis is feature engineering while it severely affects time complexity and accuracy of sentiment classification. Design/methodology/approach In this paper, a feature reduction method is proposed that uses context-based knowledge as well as synset statistical knowledge. To do so, one-dimensional presentation proposed for SentiWordNet calculates statistical knowledge that involves polarity concentration and variation tendency for each synset. Feature reduction involves two phases. In the first phase, features that combine semantic and statistical similarity conditions are put in the same cluster. In the second phase, features are ranked and then the features which are given lower ranks are eliminated. The experiments are conducted by support vector machine (SVM), naive Bayes (NB), decision tree (DT) and k-nearest neighbors (KNN) algorithms to classify the vectors of the unigram and bigram features in two classes of positive or negative sentiments. Findings The results showed that the applied clustering algorithm reduces SentiWordNet synset to less than half which reduced the size of the feature vector by less than half. In addition, the accuracy of sentiment classification is improved by at least 1.5 per cent. Originality/value The presented feature reduction method is the first use of the synset clustering for feature reduction. In this paper features reduction algorithm, first aggregates the similar features into clusters then eliminates unsatisfactory cluster.

STAGNATION LAMINAR PREMIXED CH4/AIR FLAME SUBJECTED TO THE EQUIVALENCE RATIO OSCILLATIONS

2015 ◽  
Vol 76 (1) ◽  
Author(s):  
Mohd Rosdzimin Abdul Rahman ◽  
Hishashi Tomita ◽  
Takeshi Yokomori ◽  
Toshihisa Ueda
Keyword(s):  
Flow Field ◽  
Equivalence Ratio ◽  
Reduction Method ◽  
The Other ◽  
Stagnation Flow ◽  
Velocity Perturbation ◽  
Sinusoidal Shape ◽  
Premixed Laminar ◽  
Oscillation Frequencies ◽  
Flame Position

The effect of the equivalence ratio oscillation on a premixed laminar CH4/air flame motion was studied experimentally with equivalence ratio oscillation frequencies of 2 to 15 Hz at lean equivalence ratio using stagnation flow field burner. Novel oscillator does the oscillation conditions and turbulence reduction method is used to suppress the velocity perturbation. The flame position variations at 2, 5, 10 and 15 Hz oscillation frequencies were significantly small when the amplitude of the equivalence ratio oscillation was zero. On the other hand, increase in amplitudes of the equivalence ratio oscillation increased the flame position variation significantly. The flame moved in sinusoidal shape and it can be clearly seen that the flame movement’s amplitude was proportional to the amplitudes of the equivalence ratio variations. This result showed that the velocity perturbation is significantly suppressed by turbulence reduction method in the examination range.

Sign in / Sign up

Export Citation Format

Share Document