scholarly journals Music Recommendation Algorithm Based on Multidimensional Time-Series Model Analysis

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Juanjuan Shi
Keyword(s):  
Time Series ◽  
Prototype System ◽  
Music Recommendation ◽  
Dynamic Adjustment ◽  
Medium Term ◽  
Recommendation Algorithm ◽  
Multidimensional Time Series ◽  
Behavior Preference ◽  
Short Time

This paper proposes a personalized music recommendation method based on multidimensional time-series analysis, which can improve the effect of music recommendation by using user’s midterm behavior reasonably. This method uses the theme model to express each song as the probability of belonging to several hidden themes, then models the user’s behavior as multidimensional time series, and analyzes the series so as to better predict the use of music users’ behavior preference and give reasonable recommendations. Then, a music recommendation method is proposed, which integrates the long-term, medium-term, and real-time behaviors of users and considers the dynamic adjustment of the influence weight of the three behaviors so as to further improve the effect of music recommendation by adopting the advanced long short time memory (LSTM) technology. Through the implementation of the prototype system, the feasibility of the proposed method is preliminarily verified.

scholarly journals Long-Term Data Traffic Forecasting for Network Dimensioning in LTE with Short Time Series

Electronics ◽  
2021 ◽  
Vol 10 (10) ◽  
pp. 1151
Author(s):  
Carolina Gijón ◽  
Matías Toril ◽  
Salvador Luna-Ramírez ◽  
María Luisa Marí-Altozano ◽  
José María Ruiz-Avilés
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Time Series Analysis ◽  
Data Traffic ◽  
Medium Term ◽  
Short Term ◽  
Traffic Forecasting ◽  
Series Analysis ◽  
Radio Planning

Network dimensioning is a critical task in current mobile networks, as any failure in this process leads to degraded user experience or unnecessary upgrades of network resources. For this purpose, radio planning tools often predict monthly busy-hour data traffic to detect capacity bottlenecks in advance. Supervised Learning (SL) arises as a promising solution to improve predictions obtained with legacy approaches. Previous works have shown that deep learning outperforms classical time series analysis when predicting data traffic in cellular networks in the short term (seconds/minutes) and medium term (hours/days) from long historical data series. However, long-term forecasting (several months horizon) performed in radio planning tools relies on short and noisy time series, thus requiring a separate analysis. In this work, we present the first study comparing SL and time series analysis approaches to predict monthly busy-hour data traffic on a cell basis in a live LTE network. To this end, an extensive dataset is collected, comprising data traffic per cell for a whole country during 30 months. The considered methods include Random Forest, different Neural Networks, Support Vector Regression, Seasonal Auto Regressive Integrated Moving Average and Additive Holt–Winters. Results show that SL models outperform time series approaches, while reducing data storage capacity requirements. More importantly, unlike in short-term and medium-term traffic forecasting, non-deep SL approaches are competitive with deep learning while being more computationally efficient.

scholarly journals Novel FTLRNN with Gamma Memory for Short-Term and Long-Term Predictions of Chaotic Time Series

2009 ◽  
Vol 2009 ◽  
pp. 1-21
Author(s):  
Sanjay L. Badjate ◽  
Sanjay V. Dudul
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Mean Squared Error ◽  
Chaotic Time Series ◽  
Nonstationary Time Series ◽  
Medium Term ◽  
Short Term ◽  
Squared Error ◽  
Wide Range

Multistep ahead prediction of a chaotic time series is a difficult task that has attracted increasing interest in the recent years. The interest in this work is the development of nonlinear neural network models for the purpose of building multistep chaotic time series prediction. In the literature there is a wide range of different approaches but their success depends on the predicting performance of the individual methods. Also the most popular neural models are based on the statistical and traditional feed forward neural networks. But it is seen that this kind of neural model may present some disadvantages when long-term prediction is required. In this paper focused time-lagged recurrent neural network (FTLRNN) model with gamma memory is developed for different prediction horizons. It is observed that this predictor performs remarkably well for short-term predictions as well as medium-term predictions. For coupled partial differential equations generated chaotic time series such as Mackey Glass and Duffing, FTLRNN-based predictor performs consistently well for different depths of predictions ranging from short term to long term, with only slight deterioration after k is increased beyond 50. For real-world highly complex and nonstationary time series like Sunspots and Laser, though the proposed predictor does perform reasonably for short term and medium-term predictions, its prediction ability drops for long term ahead prediction. However, still this is the best possible prediction results considering the facts that these are nonstationary time series. As a matter of fact, no other NN configuration can match the performance of FTLRNN model. The authors experimented the performance of this FTLRNN model on predicting the dynamic behavior of typical Chaotic Mackey-Glass time series, Duffing time series, and two real-time chaotic time series such as monthly sunspots and laser. Static multi layer perceptron (MLP) model is also attempted and compared against the proposed model on the performance measures like mean squared error (MSE), Normalized mean squared error (NMSE), and Correlation Coefficient (r). The standard back-propagation algorithm with momentum term has been used for both the models.

scholarly journals A comparative study of boar semen extenders with different proposed preservation times and their effect on semen quality and fertility

2016 ◽  
Vol 85 (1) ◽  
pp. 23-31 ◽  
Author(s):  
Marina Anastasia Karageorgiou ◽  
Georgios Tsousis ◽  
Constantin M. Boscos ◽  
Eleni D. Tzika ◽  
Panagiotis D. Tassis ◽  
...  
Keyword(s):  
Artificial Insemination ◽  
Semen Quality ◽  
Medium Term ◽  
Semen Collection ◽  
Chromatin Integrity ◽  
Boar Semen ◽  
Short Time

The present study compared the quality characteristics of boar semen diluted with three extenders of different proposed preservation times (short-term, medium-term and long-term). A part of extended semen was used for artificial insemination on the farm (30 sows/extender), while the remaining part was stored for three days (16–18 °C). Stored and used semen was also laboratory assessed at insemination time, on days 1 and 2 after the collection (day 0). The long-term extender was used for a short time, within 2 days from semen collection, with the aim to investigate a possible advantage over the others regarding laboratory or farm fertility indicators at the beginning of the preservation time. Viability, motility, kinetic indicators, morphology and DNA fragmentation were estimated. The results showed reduced viability, higher values for most of the kinetics, and higher immotile spermatozoa from day 1 to day 2 in all extenders; however, the long-term extender was superior compared to the other two on both days. With regard to morphology and chromatin integrity, the percentage of abnormal and fragmented spermatozoa increased on day 2 compared to day 1 for all of the extenders. However, based on the farrowing rate and the number of piglets born alive after the application of conventional artificial insemination within 2 days from semen collection/dilution, it was found that the medium-term diluents were more effective. In conclusion, it seems that the in vivo fertilization process involves more factors than simply the quality of laboratory evaluated sperm indicators, warranting further research.

scholarly journals Prediction of Financial Time Series Based on LSTM Using Wavelet Transform and Singular Spectrum Analysis

2021 ◽  
Vol 2021 ◽  
pp. 1-13
Author(s):  
Qi Tang ◽  
Ruchen Shi ◽  
Tongmei Fan ◽  
Yidan Ma ◽  
Jingyan Huang
Keyword(s):  
Time Series ◽  
Wavelet Transform ◽  
Financial Time Series ◽  
Singular Spectrum Analysis ◽  
Percentage Error ◽  
Medium Term ◽  
Short Term ◽  
Financial Time ◽  
Closing Price

In order to further overcome the difficulties of the existing models in dealing with the nonstationary and nonlinear characteristics of high-frequency financial time series data, especially their weak generalization ability, this paper proposes an ensemble method based on data denoising methods, including the wavelet transform (WT) and singular spectrum analysis (SSA), and long-term short-term memory neural network (LSTM) to build a data prediction model. The financial time series is decomposed and reconstructed by WT and SSA to denoise. Under the condition of denoising, the smooth sequence with effective information is reconstructed. The smoothing sequence is introduced into LSTM and the predicted value is obtained. With the Dow Jones industrial average index (DJIA) as the research object, the closing price of the DJIA every five minutes is divided into short term (1 hour), medium term (3 hours), and long term (6 hours), respectively. Based on root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and absolute percentage error standard deviation (SDAPE), the experimental results show that in the short term, medium term, and long term, data denoising can greatly improve the stability of the prediction and can effectively improve the generalization ability of LSTM prediction model. As WT and SSA can extract useful information from the original sequence and avoid overfitting, the hybrid model can better grasp the sequence pattern of the closing price of the DJIA.

Modeling Human Dynamics by Means of Cointegration Methodology

Methodology ◽  
2008 ◽  
Vol 4 (3) ◽  
pp. 113-131 ◽  
Author(s):  
Esther Stroe-Kunold ◽  
Joachim Werner
Keyword(s):  
Time Series ◽  
Adjustment Coefficient ◽  
Psychological Processes ◽  
Dynamic Adjustment ◽  
Short Term ◽  
Multivariate System ◽  
Human Dynamics ◽  
Integrated Time Series ◽  
Research Situation

The cointegration approach, developed by Engle and Granger (1987 ), allows treating nonstationary (i.e., integrated) time series as a multivariate system if their linear combination is stationary. Various psychological processes are often integrated but interconnected. Thus, cointegration methodology offers flexible techniques for modeling human dynamics. This paper introduces cointegration methodology to psychological researchers. After a brief introduction to time-series analysis, we summarize cointegration techniques and describe the so-called cointegration vector (β) and adjustment coefficient (α), the main parameters of cointegrated systems. We show that β and α are flexible tools for modeling long-term equilibrium as well as short-term dynamic adjustment within the multivariate system. An empirical example demonstrates advantages of cointegration methods in a typical research situation. Possible domains of application within the scope of psychology are discussed.

scholarly journals GENERALIZED APPROACH TO HURST EXPONENT ESTIMATING BY TIME SERIES

2018 ◽  
Vol 8 (1) ◽  
pp. 28-31 ◽  
Author(s):  
Lyudmyla Kirichenko ◽  
Tamara Radivilova ◽  
Vitalii Bulakh
Keyword(s):  
Time Series ◽  
Fractal Analysis ◽  
Hurst Exponent ◽  
Hypothesis Test ◽  
Interval Estimation ◽  
Stationary Time Series ◽  
Similar Time ◽  
Self Similar ◽  
Short Time

This paper presents a generalized approach to the fractal analysis of self-similar random processes by short time series. Several stages of the fractal analysis are proposed. Preliminary time series analysis includes the removal of short-term dependence, the identification of true long-term dependence and hypothesis test on the existence of a self-similarity property. Methods of unbiased interval estimation of the Hurst exponent in cases of stationary and non-stationary time series are discussed. Methods of estimate refinement are proposed. This approach is applicable to the study of self-similar time series of different nature.

Quality Monitoring and Prognostic of Electronics Using Multidimensional Time Series Method

2012 ◽  
Vol 190-191 ◽  
pp. 1029-1032
Author(s):  
Bo Wan ◽  
Li Wang ◽  
Gui Cui Fu
Keyword(s):  
Time Series ◽  
Time Series Analysis ◽  
Quality Monitoring ◽  
One Dimensional ◽  
Series Analysis ◽  
Degradation Data ◽  
Multidimensional Time Series ◽  
Time Series Method

This paper presents a quality monitoring and prognostic method to evaluating quality of electronics through monitoring degradation path. Electronics multiple performance parameter degradation data are treated as multidimensional time series and described using multidimensional time series model to take into account implements of stochastic nature of environmental variables and to predict long-term trend of performance degradation. A degradation test is processed for certain electronics and three kinds of performance parameters degradation data are monitored for prognostics. A comparison between the predicted degradation path using multidimensional time series analysis, the predicted degradation path using one-dimensional time series analysis and the real degradation path of the electronics is processed and the results show that the degradation path prediction using the suggested method is more effective than one-dimensional time series analysis.

Modelling of overflow emissions in flanders

1998 ◽  
Vol 38 (10) ◽  
pp. 41-48 ◽  
Author(s):  
G. Vaes ◽  
J. Berlamont
Keyword(s):  
Time Series ◽  
Storm Events ◽  
Short Time Series ◽  
Hydrodynamic Simulations ◽  
Sewer Systems ◽  
Linear Behaviour ◽  
Long Time ◽  
Short Time ◽  
Different Characteristics

Ideally, for emission calculations long term hydrodynamic simulations should be performed, but this requires long calculation times. Simplifications are consequently necessary. Due to the non-linear behaviour of sewer systems, hydrodynamic simulations using single storm events often will not lead to a good probability estimation of the overflow emissions. Simplified models using long time simulations give better results if they are well calibrated. To increase the accuracy hydrodynamic simulations with short time series can be used. The short time series are selected from the long time historical rainfall series using a simplified model. To test the accuracy of these three methods, hydrodynamic long term simulations were performed for several (small) sewer systems with different characteristics to compare with.

Sign in / Sign up

Export Citation Format

Share Document