Prediction Error Rate Research Articles

Models vetted against prediction error and parameter sensitivity standards can credibly evaluate ecosystem management options

A new standard for assessing model credibility is developed. This standard consists of parameter estimation, prediction error assessment, and a parameter sensitivity analysis that is driven by outside individuals who are skeptical of the model’s credibility (hereafter, skeptics). Ecological/environmental models that have a one-step-ahead prediction error rate that is better than naive forecasting — and are not excessively sensitive to small changes in their parameter values are said here to be vetted. A procedure is described that can perform this assessment on any model being evaluated for possible participation in an ecosystem management decision. Uncertainty surrounding the model’s ability to predict future values of its output variables and in the estimates of all of its parameters should be part of any effort to vett a model. The vetting procedure described herein, Prediction Error Rate-Deterministic Sensitivity Analysis (PER-DSA), incorporates these two aspects of model uncertainty. DSA in particular, requires participation by skeptics and is the reason why a successful DSA gives a model sufficient credibility to have a voice in ecosystem management decision making. But these models need to be stochastic and represent the mechanistic processes of the system being modeled. For such models, performing a PER-DSA can be computationally expensive. A cluster computing algorithm to speed-up these computations is described as one way to answer this challenge. This new standard is illustrated through a PER-DSA of a population dynamics model of South African rhinoceros (Ceratotherium simum simum).

IMPLEMENTASI MACHINE LEARNING ALGORITMA NEURAL NETWORK DALAM MEMPREDIKSI HARGA SAHAM MENGGUNAKAN RAPIDMINER

The research is aimed at implementing the Neural Network algorithm in predicting stock prices using RapidMiner. Historical stock price data is analyzed to train neural network models in identifying patterns and trends that can be used to predict future stock prices. The research phases include data collection, data preprocessing, Neural Network model development, model training, and model performance evaluation. Data used include daily closing prices, trade volumes, and other technical indicators. The data is processed through several stages of preprocesing, including normalization and filling of missing values, before being used to train Neural network models. The model was built with several layers and the number of neurons adjusted based on early experiments to find the optimal configuration. Models are trained using datasets that are divided into training sets and testing sets. The predicted result is compared to the actual value to evaluate the performance of the model. The results showed that the Neural Network model was able to predict stock prices with a fairly high accuracy rate, with an RMSE of 0.989 indicating a low prediction error rate. The research concludes that the Neural Network algorithm implemented in RapidMiner is effective in predicting stock prices and has significant potential to be used in investment decision-making. The use of RapidMiner as a data processing tool has proven to be efficient and facilitating the development and evaluation of predictive models. Keywords: Neural Network, Stock Price Prediction, RapidMiner, Machine Learning, Investment

Prediction of permeability of various geotechnical materials under different temperatures based on physical characteristics and machine learning

Accurately and efficiently predicting permeability at different temperatures is crucial in the fuel science, as it enables optimization of production efficiency, and ensures safety and stability during storage and transportation. There remains a deficiency in predictive methods using comprehensive and reasonable training datasets and validating their applicability across different materials, particularly when considering stress and temperature. This study presents a machine learning-based method to predict permeability, incorporating physical characteristics such as microscopic image features and damage factors. The image sample data augmentation method was improved for a sufficient and meaningful training dataset, and gas permeability of sandstone under different pressure conditions after various temperature treatments were obtained. After summarizing the variation patterns of each characteristic with temperature, the necessity of using selected predictive features was validated, and sample sets were constructed. We trained an extreme learning machine neural network for sandstone’s permeability under different conditions, and the relative error was compared with other representative machine learning methods. Subsequently, the same data acquisition and prediction process were applied to granite and bentonite to validate the proposed method's applicability. The findings indicate that the proposed predictive method can stably provide accurate gas permeability predictions for various geomaterials, with relatively small prediction error rates (sandstone-4.1782%, granite-4.1782%, bentonite-3.2479%). Moreover, the stability of the relative error is steadily better than other representative machine learning methods. This approach achieves an effective prediction of gas permeability under various conditions, and can potentially simulate the evolution of gas permeability for fuel sector.

Thrust ball bearing fault identification from visualized data using optimized deep network

Bearings are the most common components employed in machine parts. The bearings' movement facilitates the smooth motion. They also help with friction reduction. The faults in bearings are often caused by tribological parameters. Various methods have been developed to identify faults in bearings, but they often fail to predict these accurately. This work has concentrated on designing an effective fault-recognizing model. Therefore, a framework, the Zebra-based Radial Basis Prediction Mechanism (ZbRBPM), was proposed in this work. Initially, the bearing datasets are collected, and mineral oil (MO) lubrication is added to minimize wear and friction. The primary goal of the work is to detect and classify the fault in the thrust ball bearing. The bearing vibration data included a normal vibration signal, a ball fault, and inner and outer race faults. Hence, Zebra fitness is enhanced for the optimization of tribological parameters and fault identification. The proposed model is executed in the MATLAB system. Finally, performance criteria like accuracy, precision, recall, F-score, error rate, computation time, speed, specific wear rate, friction torque, and energy consumption are validated. The performance of the proposed ZbRBPM gains better accuracy rate as 99.5 %, 99 % of precision and f-score and 99.4 % of recall. Also it significantly reduced the prediction error rate into 0.5 % with lower computation time and very low wear and friction.

Prediction Of Student Achievement Using Artificial Neural Network And Support Vector Regression At SMK TELKOM Lampung

The analysis of student performance is crucial in vocational schools because it helps identify the challenges students face in preparing themselves for the workforce. By integrating data mining techniques such as Artificial Neural Networks (ANN), educators can enhance their understanding of factors that improve student learning outcomes. An artificial neural network (ANN) is composed of interconnected artificial neurons that can learn from input data and make complex predictions, including academic achievements. The structure and function of the human brain inspire ANN. This study compares the effec- tiveness of the artificial neural network (ANN) method with other methodologies, such as support vector regression (SVR), to predict student achievement at SMK Telkom Lampung. Primary data collected from SMK Telkom Lampung includes 4939 examples with 550 cases, 26 features, and 4 meta-attributes. Performance evaluation involves metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and coefficient of determination (R2). The coefficient of determination (R2) value of the Neural Network at 0.001 is higher than the R2 value of SVR, which reaches -0.036. Research find- ings indicate that the Artificial Neural Network model slightly outperforms the Support Vector Regression model, with lower prediction error rates and better ability to explain data variability.

Intelligent detection method with 3D ranging for external force damage monitoring of power transmission lines

The rapid development of smart grid requires more and more reliable power supply. The external force damage to transmission lines can significantly impact the safety and stability of the power grid, potentially causing electric shock accidents. Existing methods for monitoring external force damage in transmission corridors lack the ability to provide effective warnings based on the real-time distance between potential hazards and power lines. Additionally, limited computational power and storage capacity at edge terminals restrict the efficient deployment of high-precision (high complexity) visual algorithms. This study presents, for the first time, a lightweight intelligent detection method integrating detection and three-dimensional (3D) ranging. A regression loss optimized for small objects is introduced to compensate for the shortcomings of lightweight networks in detection accuracy. Simultaneously, based on the influence of convolutions in various modules of the baseline model on performance, lightweight improvements are made to the detector architecture using Omni-Dimensional Dynamic Convolution and Distribution Shifting Convolution. Finally, a 3D ranging module is integrated into the detector, involving operations such as 2D3D information matching and back-projection transformation. This method innovatively achieves automated ranging and hierarchical warning. Its effectiveness is validated in transmission corridor scenarios under various weather conditions and surveillance video. The results demonstrate that our method outperforms other algorithms in hazard detection accuracy and lightweight performance. Moreover, the distance prediction error rate is below 1.6%. The hierarchical warning solutions can be applied to more scenarios.

Creating formative HRA dependency models using the HRA dependency idioms and SACADA data, Part II: Model quantification

Human reliability analysis (HRA) identifies the causal factors impacting the occurrence of human failure events and quantifies the human failure event probabilities based on those causal factors, and requires understanding the dependency structures that exist between failure events and causal factors. Many HRA methods incorporate the dependency framework established in the Technique for Human Error Rate Prediction (THERP), which uses simple multipliers on human error probabilities (HEPs) resulting from considering only a few factors. Accordingly, those HRA methods have a limited ability and accuracy when characterizing dependency. This paper presents a methodology for using HRA data to quantify Bayesian networks built from the HRA dependency idioms. The result is an objective and traceable human reliability model that enforces formative, rather than summative, dependency. The results show that baseline HEPs in this model are reasonable, and the data-informed dependency is a significant improvement for the field.

Risk Management in Using Artificial Neural Networks

The article examines risks faced by banks during their lending processes and the mechanisms for managing these risks, utilizing modern statistical methods. Specifically, the study focused on the artificial neural network model as a technique of artificial intelligence that has successfully applied various classifications and discrimination tasks among institutions. A random sample of 46 institutions obtained loans from the branches of the National Bank of Algeria (BNA), Local Development Bank (BDL), Popular Credit of Algeria (CPA), and Agricultural and Rural Development Bank (BADR) in El Bayadh province, Algeria. Each of these institutions was characterized by 14 measurable variables with numerical values derived from the financial statements (balance sheets and income statements), as well as 3 qualitative non-accounting variables extracted from the loan applicants’ files (age of the institution, sector of activity (services/productive), institution status (viable/struggling). The sample of these 46 institutions was initially divided into two groups: 64% comprised financially stable institutions, and the other 36% were struggling institutions. The research checks whether the risk assessment of each of these 46 institutions using artificial neural networks will identify their institution status (viable/struggling) in the same way as it was in the base sample. The training phase recorded a prediction error rate of 0%, and the network testing phase misclassification rate was 5.6%. The overall correct classification rate for the multilayer artificial neural network was 92.9%, with a total error rate of 7.1%. The contribution rate of the non-accounting variable “sector of activity” was 100%, and the variable “age of the institution” was 94.4%. Other variables had minor percentages, underscoring the importance of qualitative variables in the classification process. Thus, the study proved that artificial neural network model is an effective model for distinguishing between viable and struggling institutions, significantly contributing to banking risk management.

Calculation of the minimal clinically important difference in operated patients with adult spine deformity: advantages of the ROC method and significance of prevalence in threshold selection.

The Minimal Clinically Important Difference (MCID) is crucial to evaluate management outcomes, but different thresholds have been obtained in different works. Part of this variability is due to measurement error and influence of the database, both essential for calculating the MCID. The aim of this study was to introduce the association of the ROC method in the anchor-based MCID calculation for ODI, SRS-22r, and SF-36, to objectively set the threshold for the anchor-based MCID in an adult spine deformity (ASD) population. Multicentric study based on a prospective database of consecutively operated ASD patients. An anchor question was used to assess patients' quality of life after surgery. Different approaches were used to calculate the MCID and then compared: SEM (Standard Error of Measurement), MDC (Minimal Detectable Change), and anchor-based MCID with ROC method. 516 patients were included. Those who responded with 6 and 7 to the anchor question were considered improved. The MCID ranges obtained with the ROC method exhibited the lowest variability. Prediction error rates ranged from 31% (SRS-22r) to 41% (SF-36 MCS). The MCID ranges spanned between 12 and 15 for ODI, 0.6 and 0.73 for SRS-22r, 6.62 and 7.41 for SF-36 PCS, and between 2.69 and 5.63 for SF-36 MCS. The ROC method proposes an MCID range with error rate, and can objectively determine the threshold for distinguishing improved and non-improved patients. As the MCID correlates with the utilized database and error of measurement, each study should compute its own MCID for each PROM to allow comparison among different publications. II.

A deep neural network algorithm-based approach for predicting recovery period of accidents according to construction scale

Despite ongoing safety efforts, construction sites experience a concerningly high accident rate. Notwithstanding that policies and research to reduce the risk of accidents in the construction industry have been active for a long time, the accident rate in the construction industry is considerably higher than in other industries. This trend may likely be further exacerbated by the rapid growth of large-scale construction projects driven by urban population expansion. Consequently, accurately predicting recovery periods of accidents at construction sites in advance and proactively investing in measures to mitigate them is critical for efficiently managing construction projects. Therefore, the purpose of this study is to propose a framework for developing accident prediction models based on the Deep Neural Network (DNN) algorithm according to the scale of the construction site. This study suggests DNN models and applies the DNN for each construction site scale to predict accident recovery periods. The model performance and accuracy were evaluated using mean absolute error (MAE) and root-mean-square error (RMSE) and compared with the widely used multiple regression analysis model. As a result of model comparison, the DNN models showed a lower prediction error rate than the regression analysis models for both small-to-medium and large construction sites. The findings and framework of this study can be applied as the opening stage of accident risk assessment using deep learning techniques, and the introduction of deep learning technology to safety management according to the scale of the construction site is provided as a guideline.

Optimizing injection molding parameters to minimize and prediction potential for flashing defects

The injection molding process is a manufacturing process that can produce products in a short time in large quantities, in the injection molding process the factor of setting process parameters plays a significant role in product quality, so it requires special treatment. The purpose of this study is to find the optimal parameters in the injection moulding process of yogurth container lid with polypropylene material, so that the process can reduce the incidence of flashing defects that result in the emergence of initial waste in the industrial environment. The method used in this research is to create a Response Surface Methodology Box-Behnken Design (RSM - BBD) optimization model and an Artificial Neural Network (ANN) model approach in analyzing optimal parameters and predicting the appearance of flashing defects in a designed cycle. The results obtained from this research are the optimal parameters from the RSM and ANN model recommendations, namely the clamping force setting of 70 tons, holding time 0.1 seconds, and holding pressure. The ANN model provides the highest level of prediction accuracy with an R2 value of 100% and a prediction error rate of 7.9689E-09. In comparison, the RSM model obtains a prediction accuracy level with an R2 of 71% with an error rate of 0.24315.

Two-Line Element Outlier and Space Event Detection Method Based on Multi-Strategy Genetic Algorithm

The detection of two-line element (TLE) outliers and space events play a crucial role in enhancing spatial situational awareness. Therefore, this paper addresses the issue of TLE outlier detection methods that often overlook the mutual influence of multiple factors. Hence, a Multivariate Gaussian Mixture Model (MGMM) is introduced to consider the interdependencies among various indicators. Additionally, a Multi-strategy Genetic Algorithm (MGA) is employed to adjust the complexity of the MGMM, allowing it to accurately learn the actual distribution of TLE data. Initially, the proposed method applies probabilistic fits to the predicted error rate changes for both the TLE semi-major axis and the orbital inclination. Chaos initialization, a posterior probability penalty, and local optimization iterations are subsequently integrated into the genetic algorithm. These enhancements aim to estimate the MGMM parameters, addressing issues related to poor robustness and the susceptibility of the MGMM to converge to local optima. The algorithm’s effectiveness is validated using TLE data from typical space targets. The results demonstrate that the optimized algorithm can efficiently detect outliers and maneuver events within complex TLE data. Notably, the comprehensive detection performance index, measured, using the F1 score, improved by 15.9% compared to the Gaussian mixture model. This significant improvement underscores the importance of the proposed method in bolstering the security of complex space environments.

The application of graphic language personalized emotion in graphic design

Emotion Recognition is the experience of attitude in graphic language expression and composition. People use both verbal and non-verbal behaviours to communicate their emotions. Visual communication and graphic design are always evolving to meet the demands of an increasingly affluent and culturally conscious populace. When graphic designing works, designers should consider their own opinions about related works from the audience's or customer's standpoint so that the emotion between them can resonate. Hence, this study proposes a personalized emotion recognition framework based on convolutional neural networks (PERF-CNN) to create visual content for graphic design. Graphic designers prioritize the logic of showing objects in interactive designs and use visual hierarchy and page layout approaches to respond to users' demands via typography and imagery. This ensures that the user experience is maximized. This research identifies three tiers of emotional thinking: expressive signal, emotional experience, and emotional infiltration, all of which affect graphic design. This article explores the subject of graphic design language and its ways of emotional recognition, as well as the relationship between graphic images, shapes, and feelings. CNN initially extracted expressive features from the user's face images and the poster's visual information. The clustering process categorizes the poster or advertisement images into positive, negative, and neutral classes. Research and applications of graphic design language benefit from the proposed method's experimental results, demonstrating that it outperforms conventional classification approaches in the dataset. In comparison to other popular models, the experimental results demonstrate that the proposed PERF-CNN model improves each of the following: classification accuracy (97.4 %), interaction ratio (95.6 %), emotion recognition ratio (98.9 %), rate of influence of pattern and colour features (94.4 %), and prediction error rate (6.5 %).

A novel flow rate measurement method for fire hose based on vibration signal and neural network

During fire incidents, it is imperative to monitor the flow rate of the fire hoses. Due to the constraints imposed by the fire scene, vibration-based flow measurement technique has shown great potential. However, usage of single frequency or statistic feature to predict flow rate is prone to influenced by outer noise. In this study, we constructed a dataset for fire-hose flow rate prediction task with a flow rate range of 0.35 L/s to 20.00 L/s, and sample frequency of 50 kHz. On that basis, a new variant of residual network, i.e., an RBU-STMS has been developed, integrating soft thresholding and feature pyramid mechanism. Evaluation result on the dataset has proven the effectiveness of the developed model which yields improvement of 3.85 % and 4.05 % by incorporating the two mechanisms individually and an overall improvement of 5.00 % by combining both mechanisms, compared to the conventional residual network. Furthermore, the proposed model exhibits a considerable tolerance to noisy inputs, with a predicted error rate of only 8.06 % when subjected to a 10 % standard deviation of random noise in the input, remaining below the 10 % threshold.

Streamflow prediction using support vector regression machine learning model for Tehri Dam

Accurate and reliable streamflow prediction is critical for optimising water resource management, reservoir flood operations, watershed management, and urban water management. Many researchers have published on streamflow prediction using techniques like Rainfall-Runoff modelling, Time series Models, Data-driven models, Artificial intelligence, etc. Still, there needs to be generalised method practise in the real world. The resolution of this issue lies in selecting different methods for a particular study area. This paper uses the Support vector regression machine learning model to predict the streamflow for the Tehri Dam, Uttarakhand, India, at the Daily and Ten Daily time steps. Two cases are considered in predicting daily and ten daily time steps. The first case includes four input variables: Discharge, Rainfall, Temperature, and Snow cover area. The second case comprises only three input variables: Rainfall, Temperature, and Snow cover area. Radial Kernel is used to overcome the space complexity in the datasets. The K-fold cross-validation is suitable for prediction as it averages the prediction error rate after evaluating the SVR model’s performance on various subsets of the training data. The streamflow data for daily and ten daily time steps have been collected from 2006 to 2020. The calibration period is from 2006 to 2016, and the validation period is from 2017 to 2020. Nash Sutcliffe Efficiency (NSE) and Coefficient of determination (R2) are used as the accuracy indicator in this manuscript. The lag has been observed in the daily prediction time series when three input variables are considered. For other scenarios, the respective model shows excellent results at both the temporal scale and the parametres, which play a vital role in prediction. The study also enhances the effect on the potential use of input parametres in the machine learning model.

ALTA: a simple nutritional prognostic score for patients with hepatitis B virus-related acute-on-chronic liver failure.

Malnutrition, despite being a common complication, is often neglected in patients with hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF). The objective of this study was to develop a simplified nutritional prognostic score to accurately predict mortality in HBV-ACLF patients. In this multicenter retrospective study, clinical data from 530 HBV-ACLF patients were used to create a new prognostic score, which was then validated in two external cohorts (n = 229 and 248). Four independent factors were significantly associated with 28-day mortality in HBV-ACLF patients, forming a novel prognostic score (ALTA score = 0.187 × age-0.849 × lymphocyte count-2.033 × total cholesterol-0.148 × albumin-0.971). Notably, the AUROC of ALTA score for 28/90-day mortality (0.950/0.967) were significantly higher than those of three other ACLF prognostic scores (COSSH-ACLF II, 0.864/0.734; MELD, 0.525/0.488; MELD-Na, 0.546/0.517; all P < 0.001), and three known nutritional scores (CONUT, 0.739/0.861; OPNI, 0.279/0.157; NRS-2002, 0.322/0.286; all P < 0.001). The prediction error rates of ALTA score for 28-day mortality were significantly lower than COSSH-ACLF II (7.3%), MELD (14.4%), MELD-Na (12.7%), CONUT (9.0%), OPNI (30.6%), and NRS2002 (34.1%) scores. Further classifying ALTA score into two strata, the hazard ratios of mortality at 28/90 days were notably increased in the high-risk groups compared to the low-risk group (15.959 and 5.740). These results were then validated in two external cohorts. ALTA, as a simplified nutritional prognostic score for HBV-ACLF, demonstrates superiority over the COSSH-ACLF II and other scores in predicting short-term mortality among HBV-ACLF patients. Therefore, it may be used to guide clinical management, particularly in primary care settings.

Tree based Regression Models for Predicting the Compressive Strength of Concrete at High Temperature

Predicting the compressive strength of concrete is a complicated process due to the heterogeneous mixture of concrete and high variable materials. Researchers have predicted the compressive strength of concrete for various mixes using soft computing models. In this research, compressive strength of concrete at high temperature with fly ash, super plasticizers, and fibre is predicted using three regression tree-based soft computing models (Random Forest, Random Tree, and Reduced-Error Pruning Tree (REP Tree)). The data used in this study is collected from the literature, and two-thirds of the total data is used for model training, while the remaining third is reserved for testing the prepared model. The model’s performance is evaluated based on scatter plots, variation plots, box plots, and prediction error rates, i.e., R, RMSE, and MAE. The results highlight the highest performance of the Random Forest model, with R of 0.9142; RMSE of 9.6285 MPa and MAE of 6.7931 MPa, outperforming the other competing models. Furthermore, the most influential parameter is determined using sensitivity analysis. Thus, the Random Forest model is the model that can be used for predicting the compressive strength of concrete at high temperatures.

Trading-Off Safety with Agility Using Deep Pose Error Estimation and Reinforcement Learning for Perception-Driven UAV Motion Planning

Navigation and planning for unmanned aerial vehicles (UAVs) based on visual-inertial sensors has been a popular research area in recent years. However, most visual sensors are prone to high error rates when exposed to disturbances such as excessive brightness and blur, which can lead to catastrophic performance drops in perception and motion planning systems. This study proposes a novel framework to address the coupled perception-planning problem in high-risk environments. This achieved by developing algorithms that can automatically adjust the agility of the UAV maneuvers based on the predicted error rate of the pose estimation system. The fundamental idea behind our work is to demonstrate that highly agile maneuvers become infeasible to execute when visual measurements are noisy. Thus, agility should be traded-off with safety to enable efficient risk management. Our study focuses on navigating a quadcopter through a sequence of gates on an unknown map, and we rely on existing deep learning methods for visual gate-pose estimation. In addition, we develop an architecture for estimating the pose error under high disturbance visual inputs. We use the estimated pose errors to train a reinforcement learning agent to tune the parameters of the motion planning algorithm to safely navigate the environment while minimizing the track completion time. Simulation results demonstrate that our proposed approach yields significantly fewer crashes and higher track completion rates compared to approaches that do not utilize reinforcement learning.

Application of Exponential Smoothing Method for Forecasting Spare Parts Inventory at Heavy Equipment Distributor Company

PT. Kobexindo Tractors Tbk holds a significant spare parts inventory to meet their customers' needs. Over the period from 2016 to 2023, the company experienced an average annual loss of Rp. 1,176,438,113, due to the inadequate analysis of spare parts demand, which serves as a reference in the procurement process. To address this issue, this research focuses on developing a model that can generate accurate forecasts for spare parts inventory, particularly Jungheinrich parts, to support appropriate management decisions in the procurement process at the company. The Exponential Smoothing method is chosen for its ability to handle data with fluctuating patterns and trends. This study will compare the Simple Exponential Smoothing, Double Exponential Smoothing, and Triple Exponential Smoothing methods. The data ratio used in this research is 70% for training data and 30% for testing data. The prototype development is conducted using the Python programming language. The research results indicate that the Holts Winter Exponential Smoothing Model with Multiplicative Seasonality and Multiplicative Trend (Triple Exponential) is the best method among others, as follows: 1) Train RSME (7.082307), a low RSME value on training data indicates that this model has a small prediction error rate on the data used for training. 2) Test MAPE (6.343268), a low MAPE value on test data indicates that this model provides fairly accurate predictions in percentage terms of the actual values. 3) Test RSME Values (23.160521), a sufficiently low RSME value on test data indicates that this model also successfully generalizes well on unseen data.

Hybrid parallel photonic reservoir computing with accelerated data processing speed

Reservoir computing is a bio-inspired approach especially suited for processing time-dependent information. In this work, we propose a hybrid reservoir computing scheme based on parallel photonic time-delay reservoirs with optical injection. The state of virtual nodes are evaluated from the combination of temporal outputs of parallel lasers. It is demonstrated that our proposed hybrid scheme enables photonic reservoir computer with shortened time interval θ and less virtual nodes number N, which could break through the limitation of short external cavity length in miniaturization of reservoir computing with photonic integrated circuits. The computational efficiency of photonic reservoir is numerically evaluated by chaotic time-series prediction task and memory capacity test. State-of-the-art speed performances reach 20 GSa/s with prediction error rate below 0.01. Our scheme bridges the areas of photonic integrated circuits and photonic information processing.