Rule-based Method Research Articles

Модель и алгоритм поддержки принятия решения по выбору продуктов для рекомендации пользователю на основе метода анализа статистической импликации

The article considers the issues of analyzing data that the consumer encounters when choosing products and services. The problem is extracting useful information that allows offering the user new products and services depending on his preferences. This problem is localized by recommender systems focused on using the data mining methods, such as classification, clustering, analysis of association rules - a machine learning method that detects relationships between variables in databases. Compared to other methods, the advantage of the association rule-based recommender method is its transparency: the method can show the user the inference mechanism used to make decisions. Association rule-based recommender systems use two measures that are widely popular and evaluate sets of elements and create sets of association rules, the support measure and the confidence measure. However, to get better recommendations, the quality of association rules and the way sentence ranking should be measured by some objective measure. There has been developed a model and a decision support algorithm for the choice of products for recommendations to the user based on the statistical implication analysis method. In the proposed solutions, support and credibility measures are used to create association rules; a measure of the intensity of the statistical implication is used to filter the set of rules and to rank the recommendations.

Ensemble strategy learning for imperfect information games

Algorithms with several paradigms (such as rule-based methods, game theory and reinforcement learning) have achieved great success in solving imperfect information games (IIGs). However, agents based on a single paradigm tend to be brittle in certain aspects due to the paradigm’s weaknesses. In this paper, we first present three base-solvers with diversified paradigms for IIGs, and then combine them to design three ensemble-solvers (including an attention ensemble-solver, a gradient ensemble-solver and an evolution ensemble-solver) to learn ensemble strategies given base-solvers’ strengths. We evaluate our methods on Leduc poker with nonstationary opponents and limited games. The results show that our ensemble strategy learning method can effectively integrate the advantages of various advanced individual algorithms and significantly outperform them.

Direct Clinical Applications of Natural Language Processing in Common Neurological Disorders: A Systematic Review (P5-4.005)

<h3>Objective:</h3> We sought to evaluate studies that applied natural language processing (NLP) to the diagnosis, prediction, or treatment of common neurological disorders. <h3>Background:</h3> NLP, a branch of artificial intelligence that analyzes unstructured language, is being increasingly utilized in medicine. However, the extent to which NLP has been studied in neurological disorders remains inconsistently characterized. <h3>Design/Methods:</h3> This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards and was registered with the Prospective Register of Systematic Reviews (PROSPERO; CRD42021228703). The search was conducted using MEDLINE and EMBASE on May 11, 2022. We included studies of NLP use in migraine, Parkinson disease, Alzheimer dementia, stroke and transient ischemic attack (TIA), epilepsy, or multiple sclerosis. We excluded non-English language, conference abstract, and review articles, as well as studies involving heterogeneous clinical populations or indirect clinical uses of NLP. Study characteristics were extracted and analyzed using descriptive statistics. <h3>Results:</h3> We identified 916 studies, of which 41 (4.5%), published between 2013 and 2022, were included in the final review. The most studied disorders were stroke and TIA (N=20 studies, 48.8%), followed by epilepsy (N=10, 24.4%), Alzheimer disease (N=6, 14.6%), and multiple sclerosis (N=5, 12.2%). We found no studies of NLP use in migraine or Parkinson disease that met our eligibility criteria. The main objective of NLP was diagnosis (N=20, 48.8%) followed by disease phenotyping (N=17, 41.5%), prognostication (N=9, 22%) and treatment (N=4, 9.8%). Eighteen (43.9%) studies used only machine learning approaches, 6 (14.6%) used only rule-based methods, and 17 (41.5%) used both. <h3>Conclusions:</h3> We found several gaps in neurological NLP research, with few to no studies in certain disorders, suggesting additional areas of inquiry. We observed that NLP was most widely studied in diagnostic applications, implying a potential role for NLP in augmenting diagnostic accuracy in settings with limited access to neurological expertise. <b>Disclosure:</b> Dr. Lefkovitz has nothing to disclose. Ms. Walsh has nothing to disclose. The institution of an immediate family member of Dr. Blank has received personal compensation in the range of $0-$499 for serving as a Consultant for Bristol Myers Squibb (Legal). The institution of Dr. Blank has received research support from American Epilepsy Foundation/Epilepsy Foundation/NORSE Foundation. The institution of Dr. Blank has received research support from NIA (Mount Sinai OAIC). Dr. Jette has received personal compensation in the range of $500-$4,999 for serving as an Editor, Associate Editor, or Editorial Advisory Board Member for ILAE Epilepsia. The institution of Dr. Jette has received research support from NIH. The institution of Dr. Jette has received research support from AES. The institution of Dr. Jette has received research support from PCORI. Dr. Kummer has received personal compensation in the range of $500-$4,999 for serving as a Consultant for MD Aware LLC. Dr. Kummer has received personal compensation in the range of $0-$499 for serving as a Consultant for Atheneum Partners. Dr. Kummer has received personal compensation in the range of $0-$499 for serving as a Consultant for AlphaSights. Dr. Kummer has received personal compensation in the range of $0-$499 for serving as a Consultant for Gehrson Lehrman Group . Dr. Kummer has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Syapse. Dr. Kummer has received personal compensation in the range of $0-$499 for serving as a Consultant for NeuraHealth. Dr. Kummer has received personal compensation in the range of $500-$4,999 for serving on a Scientific Advisory or Data Safety Monitoring board for Syntrillo. Dr. Kummer has stock in Syntrillo. The institution of Dr. Kummer has received research support from Icahn School of Medicine at Mount Sinai.

Identifying stroke-related quantified evidence from electronic health records in real-world studies

BackgroundStroke is one of the leading causes of death and disability worldwide. The National Institutes of Health Stroke Scale (NIHSS) scores in electronic health records (EHRs), which quantitatively describe patients' neurological deficits in evidence-based treatment, are crucial in stroke-related clinical investigations. However, the free-text format and lack of standardization inhibit their effective use. Automatically extracting the scale scores from the clinical free text so that its potential value in real-world studies is realized has become an important goal. ObjectiveThis study aims to develop an automated method to extract scale scores from the free text of EHRs. MethodsWe propose a two-step pipeline method to identify NIHSS items and numerical scores and validate its feasibility using a freely accessible critical care database: MIMIC-III (Medical Information Mart for Intensive Care III). First, we utilize MIMIC-III to create an annotated corpus. Then, we investigate possible machine learning methods for two subtasks, NIHSS item and score recognition and item-score relation extraction. In the evaluation, we conduct both task-specific and end-to-end evaluations and compare our method with the rule-based method using precision, recall and F1 scores as evaluation metrics. ResultsWe use all available discharge summaries of stroke cases in MIMIC-III. The annotated NIHSS corpus contains 312 cases, 2929 scale items, 2774 scores and 2733 relations. The results show that the best F1-score of our method was 0.9006, which was attained by combining BERT-BiLSTM-CRF and Random Forest, and it outperformed the rule-based method (F1-score = 0.8098). In the end-to-end task, our method could successfully recognize the item “1b level of consciousness questions”, the score “1” and their relation “(‘1b level of consciousness questions’, ‘1’, ‘has value’)” from the sentence “1b level of consciousness questions: said name = 1”, while the rule-based method could not. ConclusionsThe two-step pipeline method we propose is an effective approach to identify NIHSS items, scores and their relations. With its help, clinical investigators can easily retrieve and access structured scale data, thereby supporting stroke-related real-world studies.

Risk-aware controller for autonomous vehicles using model-based collision prediction and reinforcement learning

Autonomous Vehicles (AVs) have the potential to save millions of lives and increase the efficiency of transportation services. However, the successful deployment of AVs requires tackling multiple challenges related to modeling and certifying safety. State-of-the-art decision-making methods usually rely on end-to-end learning or imitation learning approaches, which still pose significant safety risks. Hence the necessity of risk-aware AVs that can better predict and handle dangerous situations. Furthermore, current approaches tend to lack explainability due to their reliance on end-to-end Deep Learning, where significant causal relationships are not guaranteed to be learned from data.This paper introduces a novel risk-aware framework for training AV agents using a bespoke collision prediction model and Reinforcement Learning (RL). The collision prediction model is based on Gaussian Processes and vehicle dynamics, and is used to generate the RL state vector. Using an explicit risk model increases the post-hoc explainability of the AV agent, which is vital for reaching and certifying the high safety levels required for AVs and other safety-sensitive applications. Experimental results obtained with a simulator and state-of-the-art RL algorithms show that the risk-aware RL framework decreases average collision rates by 15%, makes AVs more robust to sudden harsh braking situations, and achieves better performance in both safety and speed when compared to a standard rule-based method (the Intelligent Driver Model). Moreover, the proposed collision prediction model outperforms other models in the literature.

MelodyDiffusion: Chord-Conditioned Melody Generation Using a Transformer-Based Diffusion Model

Artificial intelligence, particularly machine learning, has begun to permeate various real-world applications and is continually being explored in automatic music generation. The approaches to music generation can be broadly divided into two categories: rule-based and data-driven methods. Rule-based approaches rely on substantial prior knowledge and may struggle to handle large datasets, whereas data-driven approaches can solve these problems and have become increasingly popular. However, data-driven approaches still face challenges such as the difficulty of considering long-distance dependencies when handling discrete-sequence data and convergence during model training. Although the diffusion model has been introduced as a generative model to solve the convergence problem in generative adversarial networks, it has not yet been applied to discrete-sequence data. This paper proposes a transformer-based diffusion model known as MelodyDiffusion to handle discrete musical data and realize chord-conditioned melody generation. MelodyDiffusion replaces the U-nets used in traditional diffusion models with transformers to consider the long-distance dependencies using attention and parallel mechanisms. Moreover, a transformer-based encoder is designed to extract contextual information from chords as a condition to guide melody generation. MelodyDiffusion can automatically generate diverse melodies based on the provided chords in practical applications. The evaluation experiments, in which Hits@k was used as a metric to evaluate the restored melodies, demonstrate that the large-scale version of MelodyDiffusion achieves an accuracy of 72.41% (k = 1).

Medical entity recognition and knowledge map relationship analysis of Chinese EMRs based on improved BiLSTM-CRF

With the development of medical informatization, a large number of patients’ electronic medical records (EMRs) have been accumulated in the hospital information system, which is characterized by multi-structured data form, diversified professional vocabulary categories and fuzzy vocabulary demarcation. Natural language processing (NLP) provides a silver lining for parsing electronic medical records, and the mainstream methods include dictionary-based methods, rule-based and statistical methods, and machine learning methods. Due to the semantic richness and structural diversity of Chinese text, Chinese EMRs analysis methods are still scarce compared with English EMRs. In order to overcome the problems of unstructured, multiple meanings of words and unclear word boundaries in text of Chinese electronic medical records, this paper proposes a medical entity recognition method based on RoFormerV2-BiLSTM-CRF fusion model, using BIO annotation method to annotate the recognized medical entities, and using knowledge graph to analyze the medical entity relationships identified in single patient medical record, multiple patient medical records respectively. The relationships between the medical entities identified in a single patient record and multiple patient records are analyzed using knowledge graphs. The experimental analysis was conducted on the expert-annotated dataset CCKS2019, and the results showed that the proposed method was effective for recognizing “Diseases and Diagnoses”, “Laboratory Tests”, “Imaging Examinations”, “Anatomical Sites”, “Drugs” and “Surgery” in the dataset. The average accuracy, recall and F1-Score of the proposed method for the identification of the six medical entities, are 84.8%, 83.5% and 83.9%, respectively, which were 5.3%, 8.6% and 6.8% higher than the traditional Word2Vec-BiLSTM-CRF model, and the existing Word2Vec-BiLSTM-CRF and BERT-BiLSTM-CRF models were iteratively trained with the same evaluation. The experimental results show that the proposed model performs better in medical entity recognition.

A Multi-objective approach for position alignment of track geometry measurements

This study aimed to develop a multi-objective approach for reducing the positional errors in geometry measurements of track as a linear asset. Accordingly, we evaluated and compared two alignment methods – recursive segment-wise peak alignment (RSPA) and modified correlation optimised warping (MCOW). Furthermore, a novel rule-based approach was introduced to avoid data loss while aligning the datasets of the measurements of linear assets. A case study was conducted to implement and assess the performance of these methods in reducing the positional errors in track geometry measurements. The results revealed that the rule-based method preserves all the single defects present in the datasets. Furthermore, RSPA outperforms MCOW when aligning peaks, whereas MCOW is more efficient when all the data points in the datasets have equal priority.

Lifex-fiber: an open tool for myofibers generation in cardiac computational models

BackgroundModeling the whole cardiac function involves the solution of several complex multi-physics and multi-scale models that are highly computationally demanding, which call for simpler yet accurate, high-performance computational tools. Despite the efforts made by several research groups, no software for whole-heart fully-coupled cardiac simulations in the scientific community has reached full maturity yet.ResultsIn this work we present texttt {life}^{texttt {x}}-fiber, an innovative tool for the generation of myocardial fibers based on Laplace-Dirichlet Rule-Based Methods, which are the essential building blocks for modeling the electrophysiological, mechanical and electromechanical cardiac function, from single-chamber to whole-heart simulations. texttt {life}^{texttt {x}}-fiber is the first publicly released module for cardiac simulations based on texttt {life}^{texttt {x}}, an open-source, high-performance Finite Element solver for multi-physics, multi-scale and multi-domain problems developed in the framework of the iHEART project, which aims at making in silico experiments easily reproducible and accessible to a wide community of users, including those with a background in medicine or bio-engineering.ConclusionsThe tool presented in this document is intended to provide the scientific community with a computational tool that incorporates general state of the art models and solvers for simulating the cardiac function within a high-performance framework that exposes a user- and developer-friendly interface. This report comes with an extensive technical and mathematical documentation to welcome new users to the core structure of texttt {life}^{texttt {x}}-fiber and to provide them with a possible approach to include the generated cardiac fibers into more sophisticated computational pipelines. In the near future, more modules will be successively published either as pre-compiled binaries for x86-64 Linux systems or as open source software.

Deep Q-Network-Based Efficient Driving Strategy for Mixed Traffic Flow with Connected and Autonomous Vehicles on Urban Expressways

With the increase in the number of automated vehicles, roads will contain a mix of automated vehicles and human-driven vehicles. At present, rule-based driving strategy control of automated vehicles in mixed traffic flow makes it difficult to obtain optimal control. Therefore, this study proposes a learning-based driving strategy for connected and autonomous vehicles under mixed traffic flow. The proposed method differs from other driving strategies in two respects. First, both the lane-change and car-following policies are included, and the Deep Q-network algorithm is utilized to train the two policies in a mixed traffic-flow environment. Second, the proposed driving strategy considers both traffic efficiency and safety when designing the reward function. Through simulation experiments, the differences in traffic efficiency and safety of this method and the rule-based method were compared and analyzed under different traffic densities and penetration rates of connected and autonomous vehicles. Simulation results show that the driving strategy improves the average velocity (by 7.02 km/h) and traffic safety (especially in high-density traffic), compared with traditional rule-based driving strategy.

High-Resolution Mapping of Seasonal Crop Pattern Using Sentinel Imagery in Mountainous Region of Nepal: A Semi-Automatic Approach

Sustainable agricultural management requires knowledge of where and when crops are grown, what they are, and for how long. However, such information is not yet available in Nepal. Remote sensing coupled with farmers’ knowledge offers a solution to fill this gap. In this study, we created a high-resolution (10 m) seasonal crop map and cropping pattern in a mountainous area of Nepal through a semi-automatic workflow using Sentinel-2 A/B time-series images coupled with farmer knowledge. We identified agricultural areas through iterative self-organizing data clustering of Sentinel imagery and topographic information using a digital elevation model automatically. This agricultural area was analyzed to develop crop calendars and to track seasonal crop dynamics using rule-based methods. Finally, we computed a pixel-level crop-intensity map. In the end our results were compared to ground-truth data collected in the field and published crop calendars, with an overall accuracy of 88% and kappa coefficient of 0.83. We found variations in crop intensity and seasonal crop extension across the study area, with higher intensity in plain areas with irrigation facilities and longer fallow cycles in dry and hilly regions. The semi-automatic workflow was successfully implemented in the heterogeneous topography and is applicable to the diverse topography of the entire country, providing crucial information for mapping and monitoring crops that is very useful for the formulation of strategic agricultural plans and food security in Nepal.

Determining human-coronavirus protein-protein interaction using machine intelligence

The Severe Acute Respiratory Syndrome CoronaVirus 2 (SARS-CoV-2) virus spread the novel CoronaVirus −19 (nCoV-19) pandemic, resulting in millions of fatalities globally. Recent research demonstrated that the Protein-Protein Interaction (PPI) between SARS-CoV-2 and human proteins is accountable for viral pathogenesis. However, many of these PPIs are poorly understood and unexplored, necessitating a more in-depth investigation to find latent yet critical interactions. This article elucidates the host-viral PPI through Machine Learning (ML) lenses and validates the biological significance of the same using web-based tools. ML classifiers are designed based on comprehensive datasets with five sequence-based features of human proteins, namely Amino Acid Composition, Pseudo Amino Acid Composition, Conjoint Triad, Dipeptide Composition, and Normalized Auto Correlation. A majority voting rule-based ensemble method composed of the Random Forest Model (RFM), AdaBoost, and Bagging technique is proposed that delivers encouraging statistical performance compared to other models employed in this work. The proposed ensemble model predicted a total of 111 possible SARS-CoV-2 human target proteins with a high likelihood factor ≥70%, validated by utilizing Gene Ontology (GO) and KEGG pathway enrichment analysis. Consequently, this research can aid in a deeper understanding of the molecular mechanisms underlying viral pathogenesis and provide clues for developing more efficient anti-COVID medications.

Learning and Deducing Temporal Orders

This paper studies how to determine temporal orders on attribute values in a set of tuples that pertain to the same entity, in the absence of complete timestamps. We propose a creator-critic framework to learn and deduce temporal orders by combining deep learning and rule-based deduction, referred to as GATE (Get the lATEst). The creator of GATE trains a ranking model via deep learning, to learn temporal orders and rank attribute values based on correlations among the attributes. The critic then validates the temporal orders learned and deduces more ranked pairs by chasing the data with currency constraints; it also provides augmented training data as feedback for the creator to improve the ranking in the next round. The process proceeds until the temporal order obtained becomes stable. Using real-life and synthetic datasets, we show that GATE is able to determine temporal orders with F -measure above 80%, improving deep learning by 7.8% and rule-based methods by 34.4%.

RUPA: A High Performance, Energy Efficient Accelerator for Rule-Based Password Generation in Heterogenous Password Recovery System

There has been a growing demand for energy efficient and high performance password recovery systems. As password generation and password validation are two integral components of any password recovery system, the former yet lags behind the latter in performance particularly for the case when the popular rule-based password generation method is applied to the heterogeneous CPU-FPGA system. In this paper, we thus present a high performance, energy efficient accelerator to speed up the rule functions in rule-based password generation. Dubbed RUPA, this proposed accelerator explores previously undiscovered computational features and memory access patterns for processing the rule functions. Specially, we show that the rule functions can be mapped to three distinct groups according to their character dependency graphs. Correspondingly, three kinds of datapath units, referred to as rule logic units, are created, and the rule functions from the same group will be processed in their shared rule logic unit. Compared with the state-of-the-art password recovery system built upon a CPU-GPU platform, the FPGA-based RUPA system achieves 5.3x speed improvement and is 33.1x more energy efficient. If RUPA is integrated into the popular password recovery tool John the Ripper (JtR), JtR's rule-based attack performance can soar by more than 48.7x.

DevTag: A Benchmark for Fingerprinting IoT Devices

Nowadays, various Internet of Things (IoT) devices, such as routers, webcams, and network printers, have been deployed across the Internet. For security and management purposes, it is important to accurately fingerprint IoT devices. In this work, we build a first benchmark called DevTag (IoT Device Tagging) for fingerprinting IoT devices. Specifically, DevTag supports retrieving packet-level features from IoT devices through two different data collections, passive monitoring, and active probing. For detecting IoT devices, DevTag integrates model-based and rule-based fingerprinting methods. For the model-based detection, we reimplemented five typical deep algorithms to infer IoT device classification models. For the rule-based detection, we generated nearly 41 117 rules in a unified format by analyzing several open-source tools. Furthermore, we conducted a systematic analysis to explore the advantages and limitations of those two methods for detecting IoT devices. Our analysis results reveal that the model-based detection has a significant advantage in distinguishing coarse-grained IoT devices (e.g., device type and vendor), while it is not suitable to detect product information as the label amount is massive. The rule-based detection is capable of extracting fine-grained device information with high precision in a short time. However, rules also suffer several inherent problems, such as multiple matching, conflicting, and overlapping issues. Finally, we implemented and distributed a prototype of DevTag working as the first benchmark for detecting IoT devices in the network community.

Correction to: Decision rule-based method in solving adjustable robust capacity expansion problem

Correction to: Decision rule-based method in solving adjustable robust capacity expansion problem

A Hybrid Approach of the Deep Learning Method and Rule-Based Method for Fault Diagnosis of Sucker Rod Pumping Wells

Accurately obtaining the working status of the sucker rod pumping wells is a challenging problem for oil production. Sensors at the polished rod collect working data to form surface dynamometer cards for fault diagnosis. A prevalent method for recognizing these cards is the convolutional neural network (CNN). However, this approach has two problems: an unbalanced dataset due to varying fault frequencies and similar dynamometer card shapes that complicate recognition. This leads to a low accuracy of fault diagnosis in practice, which is unsatisfactory. Therefore, this paper proposes a hybrid approach of the deep learning method and rule-based method for fault diagnosis of sucker rod pumping wells. Specifically, when the CNN model alone fails to achieve satisfactory accuracy in the working status, historical monitoring data of the relevant wells can be collected, and expert rules can assist CNN to improve diagnostic accuracy. By analyzing time series data of factors such as the maximum and minimum loads, the area of the dynamometer card, and the load difference, a knowledgebase of expert rules can be created. When performing fault diagnosis, both the dynamometer cards and related time series data are used as inputs. The dynamometer cards are used for the CNN model to diagnose, and the related time series data are used for expert rules to diagnose. The diagnostic results and the confidence levels of the two methods are obtained and compared. When the two diagnostic results conflict, the one with higher confidence is preserved. Out of the 2066 wells and 7 fault statuses analyzed in field applications, the hybrid approach demonstrated a 21.25% increase in fault diagnosis accuracy compared with using only the CNN model. Additionally, the overall accuracy rate of the hybrid approach exceeded 95%, indicating its high effectiveness in diagnosing faults in sucker rod pumping wells.

Demand forecasting of spare parts with regression and machine learning methods: Application in a bus fleet

Forecasting the demand of spare parts of vehicles in bus fleets is a vital issue. Vehicles must operate effectively and must have a high availability rate in the fleet. In maintenance operations, faulty parts or parts that complete their lifetime must be replaced with a new one. Spare parts needed must be in inventories with the required amount on time. In this sector, there are thousands of spare parts to manage. The maintenance and repair department must operate effectively. In order to accomplish this, accurate forecast of spare parts is required. In this study, demand forecasting was carried out with regression-based methods (multivariate linear regression, multivariate nonlinear regression, Gaussian process regression, additive regression, regression by discretion, support vector regression), rule-based methods (decision table, M5Rule), tree-based methods (random forest, M5P, Random tree, REPTree) and artificial neural networks. The forecasting model developed in this study includes critical variables such as the number of vehicles in the fleet, the number of breakdowns that cause parts to change, the number of periodic maintenance, mean time between failure and demand quantity in previous years. The application was carried out with real data of eight (2013–2020) years. 2013–2019 data was used for training and 2020 data was used for testing. In forecasts, support vector regression among regression-based methods, decision table among rule-based methods, M5P among tree-based methods gave the best results. It has been observed that the artificial neural network produced more accurate forecasts than all other methods. Artificial neural network forecasts give the highest forecast accuracy rate and the least deviation.

Decision rule-based method in solving adjustable robust capacity expansion problem

Decision rule-based method in solving adjustable robust capacity expansion problem

Research and application of park multi-source synergistic and complementary heating system and its coordinated control method

The integration of solar thermal collectors (STC) and air source heat pump (ASHP) heating systems has been widely used for building heating. However, these systems have the problems that the STC cannot fully utilize the solar energy at weak radiation, and the ASHP has insufficient output as well as low coefficient of performance (COP) under extreme cold conditions. In this study, a multi-source synergistic and complementary heating system (MSSCHS) with a heating mode selection model based on belief rule base and evidential reasoning (BRB-ER) method is designed. First, the MSSCHS composed of the STC and the ASHP as the main heat sources, and the water source heat pump as the auxiliary heat source, along with its three heating modes under different operating conditions were proposed. Then, a system heating mode selection model based on the BRB-ER method was establish to achieve mode switching. Finally, the MSSCHS with BRB-ER heating mode selection model was tested experimentally. The results show that compared with the conventional ASHP + STC heating system, the solar energy utilization rate of the MSSCHS had increased by 19.3 %, 16.6 % and 8.2 % respectively in three typical days, and the ASHP's COP had increased by 37.7 %- 55.6 % under extreme cold conditions. The high efficiency and stability of the MSSCHS were verified through operational tests during the heating period. Furthermore, by applying in different types of buildings, it is verified that the MSSCHS can meet different heating requirements and has a good applicability. It was demonstrated that the proposed MSSCHS has more efficient heating mode and more accurate mode selection method, which can provide a new research idea for building heating.