Task-independent Recognition of Communication Skills in Group Interaction Using Time-series Modeling

2021 ◽  
Vol 17 (4) ◽  
pp. 1-27
Author(s):  
Candy Olivia Mawalim ◽  
Shogo Okada ◽  
Yukiko I. Nakano
Keyword(s):  
Time Series ◽  
Communication Skills ◽  
Prediction Models ◽  
Short Term Memory ◽  
Group Discussion ◽  
Task Type ◽  
Support Vector ◽  
Time Series Modeling ◽  
Group Discussions ◽  
Multimodal Analysis

Case studies of group discussions are considered an effective way to assess communication skills (CS). This method can help researchers evaluate participants’ engagement with each other in a specific realistic context. In this article, multimodal analysis was performed to estimate CS indices using a three-task-type group discussion dataset, the MATRICS corpus. The current research investigated the effectiveness of engaging both static and time-series modeling, especially in task-independent settings. This investigation aimed to understand three main points: first, the effectiveness of time-series modeling compared to nonsequential modeling; second, multimodal analysis in a task-independent setting; and third, important differences to consider when dealing with task-dependent and task-independent settings, specifically in terms of modalities and prediction models. Several modalities were extracted (e.g., acoustics, speaking turns, linguistic-related movement, dialog tags, head motions, and face feature sets) for inferring the CS indices as a regression task. Three predictive models, including support vector regression (SVR), long short-term memory (LSTM), and an enhanced time-series model (an LSTM model with a combination of static and time-series features), were taken into account in this study. Our evaluation was conducted by using the R 2 score in a cross-validation scheme. The experimental results suggested that time-series modeling can improve the performance of multimodal analysis significantly in the task-dependent setting (with the best R 2 = 0.797 for the total CS index), with word2vec being the most prominent feature. Unfortunately, highly context-related features did not fit well with the task-independent setting. Thus, we propose an enhanced LSTM model for dealing with task-independent settings, and we successfully obtained better performance with the enhanced model than with the conventional SVR and LSTM models (the best R 2 = 0.602 for the total CS index). In other words, our study shows that a particular time-series modeling can outperform traditional nonsequential modeling for automatically estimating the CS indices of a participant in a group discussion with regard to task dependency.

scholarly journals Generation of Head Movements of a Robot Using Multimodal Features of Peer Participants in Group Discussion Conversation

2020 ◽  
Vol 4 (2) ◽  
pp. 15
Author(s):  
Hung-Hsuan Huang ◽  
Seiya Kimura ◽  
Kazuhiro Kuwabara ◽  
Toyoaki Nishida
Keyword(s):  
Communication Skills ◽  
Research Question ◽  
Group Discussion ◽  
Attentional Focus ◽  
Head Movements ◽  
Support Vector ◽  
Group Discussions ◽  
Multimodal Features ◽  
Significant Difference ◽  
Behavior Frequency

In recent years, companies have been seeking communication skills from their employees. Increasingly more companies have adopted group discussions during their recruitment process to evaluate the applicants’ communication skills. However, the opportunity to improve communication skills in group discussions is limited because of the lack of partners. To solve this issue as a long-term goal, the aim of this study is to build an autonomous robot that can participate in group discussions, so that its users can repeatedly practice with it. This robot, therefore, has to perform humanlike behaviors with which the users can interact. In this study, the focus was on the generation of two of these behaviors regarding the head of the robot. One is directing its attention to either of the following targets: the other participants or the materials placed on the table. The second is to determine the timings of the robot’s nods. These generation models are considered in three situations: when the robot is speaking, when the robot is listening, and when no participant including the robot is speaking. The research question is: whether these behaviors can be generated end-to-end from and only from the features of peer participants. This work is based on a data corpus containing 2.5 h of the discussion sessions of 10 four-person groups. Multimodal features, including the attention of other participants, voice prosody, head movements, and speech turns extracted from the corpus, were used to train support vector machine models for the generation of the two behaviors. The performances of the generation models of attentional focus were in an F-measure range between 0.4 and 0.6. The nodding model had an accuracy of approximately 0.65. Both experiments were conducted in the setting of leave-one-subject-out cross validation. To measure the perceived naturalness of the generated behaviors, a subject experiment was conducted. In the experiment, the proposed models were compared. They were based on a data-driven method with two baselines: (1) a simple statistical model based on behavior frequency and (2) raw experimental data. The evaluation was based on the observation of video clips, in which one of the subjects was replaced by a robot performing head movements in the above-mentioned three conditions. The experimental results showed that there was no significant difference from original human behaviors in the data corpus and proved the effectiveness of the proposed models.

Forecasting the risk at infractions: an ensemble comparison of machine learning approach

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Lei Li ◽  
Desheng Wu
Keyword(s):  
Machine Learning ◽  
Prediction Models ◽  
Short Term Memory ◽  
Model Performance ◽  
Large Data ◽  
Support Vector ◽  
Learning Approaches ◽  
Content Type ◽  
Day To Day Operations ◽  
Prediction Approach

PurposeThe infraction of securities regulations (ISRs) of listed firms in their day-to-day operations and management has become one of common problems. This paper proposed several machine learning approaches to forecast the risk at infractions of listed corporates to solve financial problems that are not effective and precise in supervision.Design/methodology/approachThe overall proposed research framework designed for forecasting the infractions (ISRs) include data collection and cleaning, feature engineering, data split, prediction approach application and model performance evaluation. We select Logistic Regression, Naïve Bayes, Random Forest, Support Vector Machines, Artificial Neural Network and Long Short-Term Memory Networks (LSTMs) as ISRs prediction models.FindingsThe research results show that prediction performance of proposed models with the prior infractions provides a significant improvement of the ISRs than those without prior, especially for large sample set. The results also indicate when judging whether a company has infractions, we should pay attention to novel artificial intelligence methods, previous infractions of the company, and large data sets.Originality/valueThe findings could be utilized to address the problems of identifying listed corporates' ISRs at hand to a certain degree. Overall, results elucidate the value of the prior infraction of securities regulations (ISRs). This shows the importance of including more data sources when constructing distress models and not only focus on building increasingly more complex models on the same data. This is also beneficial to the regulatory authorities.

scholarly journals Spatiotemporal Traffic Flow Prediction with KNN and LSTM

2019 ◽  
Vol 2019 ◽  
pp. 1-10 ◽  
Author(s):  
Xianglong Luo ◽  
Danyang Li ◽  
Yu Yang ◽  
Shengrui Zhang
Keyword(s):  
Support Vector Regression ◽  
Traffic Flow ◽  
Prediction Models ◽  
Short Term Memory ◽  
Transportation Systems ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Traffic Flow Prediction ◽  
Flow Prediction ◽  
Proposed Model

The traffic flow prediction is becoming increasingly crucial in Intelligent Transportation Systems. Accurate prediction result is the precondition of traffic guidance, management, and control. To improve the prediction accuracy, a spatiotemporal traffic flow prediction method is proposed combined with k-nearest neighbor (KNN) and long short-term memory network (LSTM), which is called KNN-LSTM model in this paper. KNN is used to select mostly related neighboring stations with the test station and capture spatial features of traffic flow. LSTM is utilized to mine temporal variability of traffic flow, and a two-layer LSTM network is applied to predict traffic flow respectively in selected stations. The final prediction results are obtained by result-level fusion with rank-exponent weighting method. The prediction performance is evaluated with real-time traffic flow data provided by the Transportation Research Data Lab (TDRL) at the University of Minnesota Duluth (UMD) Data Center. Experimental results indicate that the proposed model can achieve a better performance compared with well-known prediction models including autoregressive integrated moving average (ARIMA), support vector regression (SVR), wavelet neural network (WNN), deep belief networks combined with support vector regression (DBN-SVR), and LSTM models, and the proposed model can achieve on average 12.59% accuracy improvement.

A Traffic Prediction Method Based on the Hybrid Model of CLWT

2015 ◽  
Vol 713-715 ◽  
pp. 1564-1569
Author(s):  
Jin Long Fei ◽  
Wei Lin ◽  
Tao Han ◽  
Yue Fei Zhu
Keyword(s):  
Time Series ◽  
Hybrid Model ◽  
Network Traffic ◽  
Prediction Models ◽  
Prediction Method ◽  
Support Vector ◽  
Prediction Methods ◽  
Recurrent Nerve ◽  
Nerve Network ◽  
Traffic Time Series

Current prediction models for network traffic cannot accurately depict the multi-properties of the Internet traffic. This paper proposes a wavelet-based hybrid model prediction method for network traffic called CLWT model and proposes a prediction method for traffic based on this model. The traffic time series can be rapidly decomposed respectively into approximate time series and detail time series with LF and HF response. The approximate time series predicts by making use of Least Squares Support Vector Machine and proceeds error calibration by using Generalized Recurrent Nerve Network. The detail time series predict it by making use of self-adaption chaotic prediction methods after the medium-soft threshold noise reduction. Finally the prediction value of time series is got by making use of promoting wavelet reconstitution. The effectiveness for the prediction methods mentioned in the paper has been validated by simulation experiment. High prediction accuracy is obtained compared with the existing methods.

scholarly journals An Optimized Method of Wind Speed Prediction with Support Vector Machine and Genetic Algorithm

2021 ◽  
Vol 15 ◽  
pp. 212-217
Author(s):  
Guan-fa Li ◽  
Wen-sheng Zhu
Keyword(s):  
Genetic Algorithm ◽  
Time Series ◽  
Support Vector Machine ◽  
Wind Speed ◽  
Power System ◽  
Prediction Models ◽  
Optimization Method ◽  
Support Vector ◽  
Wind Speed Prediction ◽  
Speed Prediction

Due to the randomness of wind speed and direction, the output power of wind turbine also has randomness. After large-scale wind power integration, it will bring a lot of adverse effects on the power quality of the power system, and also bring difficulties to the formulation of power system dispatching plan. In order to improve the prediction accuracy, an optimized method of wind speed prediction with support vector machine and genetic algorithm is put forward. Compared with other optimization methods, the simulation results show that the optimized genetic algorithm not only has good convergence speed, but also can find more suitable parameters for data samples. When the data is updated according to time series, the optimization range of vaccine and parameters is adaptively adjusted and updated. Therefore, as a new optimization method, the optimization method has certain theoretical significance and practical application value, and can be applied to other time series prediction models.

scholarly journals COLLAPSE WARNING SYSTEM USING LSTM NEURAL NETWORKS FOR CONSTRUCTION DISASTER PREVENTION IN EXTREME WIND WEATHER

2021 ◽  
Vol 27 (4) ◽  
pp. 230-245
Author(s):  
Chih-Chiang Wei
Keyword(s):  
Neural Networks ◽  
Prediction Models ◽  
Short Term Memory ◽  
Meteorological Data ◽  
Warning System ◽  
Weather Conditions ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Strong Wind ◽  
Wind Force

Strong wind during extreme weather conditions (e.g., strong winds during typhoons) is one of the natural factors that cause the collapse of frame-type scaffolds used in façade work. This study developed an alert system for use in determining whether the scaffold structure could withstand the stress of the wind force. Conceptually, the scaffolds collapsed by the warning system developed in the study contains three modules. The first module involves the establishment of wind velocity prediction models. This study employed various deep learning and machine learning techniques, namely deep neural networks, long short-term memory neural networks, support vector regressions, random forest, and k-nearest neighbors. Then, the second module contains the analysis of wind force on the scaffolds. The third module involves the development of the scaffold collapse evaluation approach. The study area was Taichung City, Taiwan. This study collected meteorological data from the ground stations from 2012 to 2019. Results revealed that the system successfully predicted the possible collapse time for scaffolds within 1 to 6 h, and effectively issued a warning time. Overall, the warning system can provide practical warning information related to the destruction of scaffolds to construction teams in need of the information to reduce the damage risk.

scholarly journals Prognosis of Chronic Renal Syndrome by Classification and Progression Using Temporal Abstraction

2019 ◽  
Vol 8 (11) ◽  
pp. 53-58
Keyword(s):  
Genetic Algorithm ◽  
Time Series ◽  
Renal Disease ◽  
Prediction Model ◽  
Prediction Models ◽  
Series Data ◽  
Support Vector ◽  
Temporal Abstraction ◽  
Graphic Data ◽  
Lab Test

Chronic renal syndrome is defined as a progressive loss of renal function over period. Analysers have make effort in attempting to diagnosis the risk factors that may affect the retrogression of chronic renal syndrome. The motivation of this project helps to develop a prediction model for level 4 CKD patients to detect on condition that, their estimated Glomerular Filtration Rate (eGFR) stage downscale to lower than 15 ml/min/1.73 m². End phase renal disease, after six months accumulating their concluding lab test observation by assessing time affiliated aspects. Data mining algorithm along with Temporal Abstraction (TA) are confederated to reinforce CKD evolvement of prognostication models. In this work a inclusive of 112 chronic renal disease patients are composed from April 1952 to September 2011 which were extracted from the patient’s Electronic Medical Records (EMR). The information of chronic renal patients are collected in a big spatial info-graphic data. In order to analyse these info-graphic data, it is significant to detect the issues affecting CKD deterioration and hence it becomes a challenging task. To overcome this challenge, time series graph has been generated in this project work based on creatinine and albumin lab test values and reports of the time period. The presence of CKD diagnostic codes are transformed into default seven digit default format of International Classification of Disease 10 Clinical Modification (ICD 10 CM). Feature selection is performed in this work based on wrapper method using genetic algorithm. It is helpful for finding the most relevant variables for a predictive model. High Utility Sequential Rule Miner (HUSRM) is used here to address the discovery of CKD sequential rules based on sequence patterns. Temporal Abstraction (TA) techniques namely basic TA and complex TA are used in this work to analyse the status of chronic renal syndrome patients. Classification and Regression Technique (CART) along with Adaptive Boosting (AdaBoost) and Support Vector Machine Boosting (SVMBoost) are applied to develop the CKD in which the progression prediction models exhibit most accurate prediction. The results obtained from this work divulged that comprehending temporal observation forward the prognostic instances has escalated the efficacy of the instances. Finally, an evaluation metrics namely accuracy, sensitivity, specificity, positive likelihood, negative likelihood and Area Under the Curve (AUC) are helps to evaluate the performance of the prediction models which are designed and implemented in this project. Key Words: CKD, progression, time series data, genetic algorithm, sequential rules, TA classification and prediction model.

Comprehensive Data Analysis and Forecasting of Covid19 Outbreak in India (Preprint)

2020 ◽  
Author(s):  
Amit Thakur ◽  
Rajesh Singh ◽  
Anita Gehlot ◽  
Shaik Vaseem Akram ◽  
Prabin Kumar Das
Keyword(s):  
Time Series ◽  
Support Vector Machine ◽  
Linear Regression ◽  
Input Data ◽  
Good Accuracy ◽  
Prediction Models ◽  
Indian Region ◽  
Support Vector ◽  
Rate Of Growth ◽  
Good Range

BACKGROUND COVID-19 is chronic based disease which is spreading with rapid pace in the entire world. Present study addresses the situation of outbreak of the COVID-19 disease in India and estimate the rise of the cases in India. This study addresses the present health infrastructure, infected health workforce clearly with the statistics. Support Vector Machine and Linear Regression are implemented in this study for predicting the expected cases. For the purpose of modelling, the input data of number of cases is considered from the march 15th , 2020. With the input data, the two models are trained for prediction of the cases. In the end, the results show that support vector machine and linear regression are giving good accuracy for prediction. OBJECTIVE The current studies aim to analyze and estimate the developments in the near future with reference to COVID-19 in India. The research is also planned to look at the preparation level of Indian government for this outbreak. The scope of the study is narrowed to build prediction models for the Indian region and uses SVMs for prediction methods based on time series that are easily built and readable under these crucial conditions. The study does not cover coverage of a COVID-19 outbreak for any other country. METHODS Support Vector Machine and Linear Regression are implemented in this study for predicting the expected cases. For the purpose of modelling, the input data of number of cases is considered from the march 15th , 2020. With the input data, the two models are trained for prediction of the cases. In the end, the results show that support vector machine and linear regression are giving good accuracy for prediction. RESULTS 1.Considering the change, the change in slope of the both curves in the graph, it can be concluded that the trained model is giving a quite good range of accuracy. 2.The Graph shows the plot of the predicted values and actual values fed during the testing of model. Considering the change, the change in slope of the both curves in the graph, it can be concluded that the trained model is giving a quite good range of accuracy. CONCLUSIONS In conclusion, the present work emphasized on presenting observations and predictions about COVID-19 outbreaks in the Indian region. Although the rate of growth at world level is not equal to the rate of growth, the situation appears dangerous as India is heading towards exponential growth. The expected patients are reaching in millions in the next 30 days by means of two separate time series forecasting models. With regard to the poor health facilities, it is going to difficult to combat the outbreak of virus without government addressing the effective measurements. Contrast to strict lockdown, social distancing, isolation, patient testing and medical care need to implement with war base for combating the pandemic in India. The forecasting in this study are still in beginning phases as the historical data is limit for creating reliable model. That to the risen of cases in India followed from the last 10 days so the training for the model may not be accurate, however the prediction model would be enhanced from existing models, as the greater number of medical and demographic data is available.Furthermore, even if the predictions are 60-70 percent correct, then the nation will also encounter this quite hard days.

scholarly journals Development of a Three-Stage Hybrid Model by Utilizing a Two-Stage Signal Decomposition Methodology and Machine Learning Approach to Predict Monthly Runoff at Swat River Basin, Pakistan

2020 ◽  
Vol 2020 ◽  
pp. 1-19 ◽  
Author(s):  
Muhammad Sibtain ◽  
Xianshan Li ◽  
Ghulam Nabi ◽  
Muhammad Imran Azam ◽  
Hassan Bashir
Keyword(s):  
Time Series ◽  
Hybrid Model ◽  
Prediction Models ◽  
Hybrid Models ◽  
Signal Decomposition ◽  
High Frequency Component ◽  
Support Vector ◽  
Two Stage ◽  
Runoff Prediction ◽  
Swat River

Precise and reliable hydrological runoff prediction plays a significant role in the optimal management of hydropower resources. Nevertheless, the hydrological runoff practically possesses a nonlinear dynamics, and constructing appropriate runoff prediction models to deal with the nonlinearity is a challenging task. To overcome this difficulty, this paper proposes a three-stage novel hybrid model, namely, CVS (CEEMDAN-VMD-SVM), by coupling the support vector machine (SVM) with a two-stage signal decomposition methodology, combining complete ensemble empirical decomposition with additive noise (CEEMDAN) and variational mode decomposition (VMD), to obtain inclusive information of the runoff time series. Hydrological runoff data of the Swat River, Pakistan, from 1961 to 2015 were taken for prediction. CEEMDAN decomposes the runoff time series into subcomponents, and VMD performs further decomposition of the high-frequency component obtained after CEEMDAN decomposition to improve the prediction activity. Afterward, the SVM algorithm was applied to the decomposed subcomponents for the prediction purpose. Finally, four statistical indices are utilized to measure the performance of the CVS model compared with other hybrid models including CEEMDAN-VMD-MLP (multilayer perceptron), CEEMDAN-SVM, VMD-SVM, CEEMDAN-MLP, VMD-MLP, SVM, and MLP. The CVS model performs better during the training period by reducing RMSE by 71.28% and 40.06% compared with MLP and CEEDMAD-VMD-SVM models, respectively. However, during the testing period, the error reductions include RMSE by 68.37% and 35.33% compared with MLP and CEEDMAD-VMD-SVM models, respectively. The results highlight that the CVS model outperforms other models in terms of accuracy and error reduction. The research also highlights the superiority of other hybrid models over standalone in predicting the hydrological runoff. Therefore, the proposed hybrid model is applicable for the nonlinear features of runoff time series with feasibility for future planning and management of water resources.

Sign in / Sign up

Export Citation Format

Share Document