Objective To improve the accuracy and professionalism of question-answering(QA)model in traditional Chinese medicine(TCM)lung cancer by integrating large language models with structured knowledge graphs using the know...Objective To improve the accuracy and professionalism of question-answering(QA)model in traditional Chinese medicine(TCM)lung cancer by integrating large language models with structured knowledge graphs using the knowledge graph(KG)to text-enhanced retrievalaugmented generation(KG2TRAG)method.Methods The TCM lung cancer model(TCMLCM)was constructed by fine-tuning Chat-GLM2-6B on the specialized datasets Tianchi TCM,HuangDi,and ShenNong-TCM-Dataset,as well as a TCM lung cancer KG.The KG2TRAG method was applied to enhance the knowledge retrieval,which can convert KG triples into natural language text via ChatGPT-aided linearization,leveraging large language models(LLMs)for context-aware reasoning.For a comprehensive comparison,MedicalGPT,HuatuoGPT,and BenTsao were selected as the baseline models.Performance was evaluated using bilingual evaluation understudy(BLEU),recall-oriented understudy for gisting evaluation(ROUGE),accuracy,and the domain-specific TCM-LCEval metrics,with validation from TCM oncology experts assessing answer accuracy,professionalism,and usability.Results The TCMLCM model achieved the optimal performance across all metrics,including a BLEU score of 32.15%,ROUGE-L of 59.08%,and an accuracy rate of 79.68%.Notably,in the TCM-LCEval assessment specific to the field of TCM,its performance was 3%−12%higher than that of the baseline model.Expert evaluations highlighted superior performance in accuracy and professionalism.Conclusion TCMLCM can provide an innovative solution for TCM lung cancer QA,demonstrating the feasibility of integrating structured KGs with LLMs.This work advances intelligent TCM healthcare tools and lays a foundation for future AI-driven applications in traditional medicine.展开更多
Question-answering(QA)models find answers to a given question.The necessity of automatically finding answers is increasing because it is very important and challenging from the large-scale QA data sets.In this paper,w...Question-answering(QA)models find answers to a given question.The necessity of automatically finding answers is increasing because it is very important and challenging from the large-scale QA data sets.In this paper,we deal with the QA pair matching approach in QA models,which finds the most relevant question and its recommended answer for a given question.Existing studies for the approach performed on the entire dataset or datasets within a category that the question writer manually specifies.In contrast,we aim to automatically find the category to which the question belongs by employing the text classification model and to find the answer corresponding to the question within the category.Due to the text classification model,we can effectively reduce the search space for finding the answers to a given question.Therefore,the proposed model improves the accuracy of the QA matching model and significantly reduces the model inference time.Furthermore,to improve the performance of finding similar sentences in each category,we present an ensemble embedding model for sentences,improving the performance compared to the individual embedding models.Using real-world QA data sets,we evaluate the performance of the proposed QA matching model.As a result,the accuracy of our final ensemble embedding model based on the text classification model is 81.18%,which outperforms the existing models by 9.81%∼14.16%point.Moreover,in terms of the model inference speed,our model is faster than the existing models by 2.61∼5.07 times due to the effective reduction of search spaces by the text classification model.展开更多
Geological strength index(GSI)has been widely used as an input parameter in predicting the strength and deformation properties of rock masses.This study derived a series of equations to satisfy the original GSI lines ...Geological strength index(GSI)has been widely used as an input parameter in predicting the strength and deformation properties of rock masses.This study derived a series of equations to satisfy the original GSI lines on the basic GSI chart.Two axes ranging from 0 to 100 were employed for surface conditions of the discontinuities and the structure of rock mass,which are independent of the input parameters.The derived equations can analyze GSI values ranging from 0 to 100 within±5%error.The engineering dimensions(EDs)such as the slope height,tunnel width,and foundation width were used together with representative elementary volume(REV)in jointed rock mass to define scale factor(sf)from 0.2 to 1 in evaluating the rock mass structure including joint pattern.The transformation of GSI into a scaledependent parameter based on engineering scale addresses a crucial requirement in various engineering applications.The improvements proposed in this study were applied to a real slope which was close to the time of failure.The results of stability assessments show that the new proposals have sufficient capability to define rock mass quality considering EDs.展开更多
The quality control chart of fructo-oligosaccharides in milk powder was established to determine whether the detection process and results are in control state.The content of fructo-oligosaccharides in milk powder con...The quality control chart of fructo-oligosaccharides in milk powder was established to determine whether the detection process and results are in control state.The content of fructo-oligosaccharides in milk powder control samples was determined by ion chromatography,and the quality control chart of fructo-oligosaccharides was established to analyze the controlled state.The results indicate that the median of the quality control chart is 1613.14 mg/100 g,and the standard deviation is 85.57 mg/100 g.The new quality control points were evaluated and analyzed,and the precision changed,but the mean value did not change.Further F test was conducted to determine that the precision did not change significantly,indicating that the test was in a statistical control state,and the detection process,method and results were controlled.展开更多
This study aims to examine the explicit solution for calculating the Average Run Length(ARL)on the triple exponentially weighted moving average(TEWMA)control chart applied to autoregressive model(AR(p)),where AR(p)is ...This study aims to examine the explicit solution for calculating the Average Run Length(ARL)on the triple exponentially weighted moving average(TEWMA)control chart applied to autoregressive model(AR(p)),where AR(p)is an autoregressive model of order p,representing a time series with dependencies on its p previous values.Additionally,the study evaluates the accuracy of both explicit and numerical integral equation(NIE)solutions for AR(p)using the TEWMA control chart,focusing on the absolute percentage relative error.The results indicate that the explicit and approximate solutions are in close agreement.Furthermore,the study investigates the performance of exponentially weighted moving average(EWMA)and TEWMA control charts in detecting changes in the process,using the relative mean index(RMI)as a measure.The findings demonstrate that the TEWMA control chart outperforms the EWMA control chart in detecting process changes,especially when the value ofλis sufficiently large.In addition,an analysis using historical data from the SET index between January 2024 and May 2024 and historical data of global annual plastic production,the results of both data sets also emphasize the superior performance of the TEWMA control chart.展开更多
Extracting data from visually rich documents and charts using traditional methods that rely on OCR-based parsing poses multiple challenges,including layout complexity in unstructured formats,limitations in recognizing...Extracting data from visually rich documents and charts using traditional methods that rely on OCR-based parsing poses multiple challenges,including layout complexity in unstructured formats,limitations in recognizing visual elements,and the correlation between different parts of the documents,as well as domain-specific semantics.Simply extracting text is not sufficient;advanced reasoning capabilities are proving to be essential to analyze content and answer questions accurately.This paper aims to evaluate the ability of the Large Language Models(LLMs)to correctly answer questions about various types of charts,comparing their performance when using images as input versus directly parsing PDF files.To retrieve the images from the PDF,ColPali,a model leveraging state-of-the-art visual languagemodels,is used to identify the relevant page containing the appropriate chart for each question.Google’s Gemini multimodal models were used to answer a set of questions through two approaches:1)processing images derived from PDF documents and 2)directly utilizing the content of the same PDFs.Our findings underscore the limitations of traditional OCR-based approaches in visual document understanding(VrDU)and demonstrate the advantages of multimodal methods in both data extraction and reasoning tasks.Through structured benchmarking of chart question answering(CQA)across input formats,our work contributes to the advancement of chart understanding(CU)and the broader field of multimodal document analysis.Using two diverse and information-rich sources:the World Health Statistics 2024 report by theWorld Health Organisation and the Global Banking Annual Review 2024 by McKinsey&Company,we examine the performance ofmultimodal LLMs across different input modalities,comparing their effectiveness in processing charts as images versus parsing directly from PDF content.These documents were selected due to their multimodal nature,combining dense textual analysis with varied visual representations,thus presenting realistic challenges for vision-language models.This comparison is aimed at assessing how advanced models perform with different input formats and to determine if an image-based approach enhances chart comprehension in terms of accurate data extraction and reasoning capabilities.展开更多
We use transaction-level data from the Bitcoin exchange Mt.Gox,including over 1.4 million transactions from more than 45,000 traders,to investigate the role of technical chart patterns in the early Bitcoin market from...We use transaction-level data from the Bitcoin exchange Mt.Gox,including over 1.4 million transactions from more than 45,000 traders,to investigate the role of technical chart patterns in the early Bitcoin market from April 2011 to September 2013.Employing a pattern recognition algorithm,we identify hourly trading signals for five major chart patterns.Buy signals of these patterns are associated with an average increase in abnormal trading volume of more than 53%.Trades executed during buy signal periods yield significantly higher average returns than those made during non-signal periods.Traders who use chart patterns more frequently are more likely to generate right-skewed return distributions,engage in more active trading,and achieve higher average roundtrip returns.Our research suggests that chart pattern trading was a crucial tool for Mt.Gox clients,highlighting the importance of technical heuristics in shaping the dynamics in a less efficient and unregulated market environment.By leveraging a comprehensive transaction dataset from a major cryptocurrency exchange,we provide unique insights into the actual trading behavior of the first Bitcoin adopters.This sets our work apart from previous studies that mainly rely on backtesting technical strategies using publicly available price data.展开更多
BACKGROUND Restless legs syndrome(RLS)is a neurological disorder characterized by an uncontrollable urge to move the legs during rest.Standard treatments may provide short-term relief but are limited by long-term adve...BACKGROUND Restless legs syndrome(RLS)is a neurological disorder characterized by an uncontrollable urge to move the legs during rest.Standard treatments may provide short-term relief but are limited by long-term adverse effects.Herbal medicine containing Paeoniae Radix(HCP)has traditionally been used in Korean medicine to treat RLS symptoms.AIM To investigate the clinical effectiveness and safety of HCP in treating RLS.METHODS This retrospective chart review study included 15 patients with RLS who were treated with HCP between September 2017 and August 2022 at the Kyung Hee University Korean Medicine Hospital.The demographic,disease-related,and treatment-related characteristics,as well as the impressions of each patient,were analyzed.The study also evaluated the changes in the subjective complaints of RLS symptoms,the Korean version of the International Restless Legs Scale(KIRLS)scores before and after treatment,and the adverse events after HCP administration.RESULTS Following HCP administration,14 patients(93.3%)experienced a decrease in the degree of discomfort associated with their symptoms,with an average reduction to 44.7±35.1%compared to their discomfort levels before taking the drug.After treatment,the mean total K-IRLS score decreased significantly from 23.00±9.75 to 13.93±10.9(P=0.005).Among the HCP,a significant reduction in the K-IRLS score was observed in the Gyejibokryeonghwan(26.0±9.7 to 10.0±9.4,P=0.028)and Jakyakgamchotang(23.2±12.6 to 6.6±4.2,P=0.042)groups.Additionally,no adverse events were reported.CONCLUSION HCP improved symptoms in RLS patients.These results suggest that HCP may be an effective and safe alternative for RLS.A randomized controlled trial is required to confirm these findings.展开更多
This study presents an innovative development of the exponentially weighted moving average(EWMA)control chart,explicitly adapted for the examination of time series data distinguished by seasonal autoregressive moving ...This study presents an innovative development of the exponentially weighted moving average(EWMA)control chart,explicitly adapted for the examination of time series data distinguished by seasonal autoregressive moving average behavior—SARMA(1,1)L under exponential white noise.Unlike previous works that rely on simplified models such as AR(1)or assume independence,this research derives for the first time an exact two-sided Average Run Length(ARL)formula for theModified EWMAchart under SARMA(1,1)L conditions,using a mathematically rigorous Fredholm integral approach.The derived formulas are validated against numerical integral equation(NIE)solutions,showing strong agreement and significantly reduced computational burden.Additionally,a performance comparison index(PCI)is introduced to assess the chart’s detection capability.Results demonstrate that the proposed method exhibits superior sensitivity to mean shifts in autocorrelated environments,outperforming existing approaches.The findings offer a new,efficient framework for real-time quality control in complex seasonal processes,with potential applications in environmental monitoring and intelligent manufacturing systems.展开更多
基金Postgraduate Research&Practice Innovation Program of Jiangsu Province(KYCX24_2145).
文摘Objective To improve the accuracy and professionalism of question-answering(QA)model in traditional Chinese medicine(TCM)lung cancer by integrating large language models with structured knowledge graphs using the knowledge graph(KG)to text-enhanced retrievalaugmented generation(KG2TRAG)method.Methods The TCM lung cancer model(TCMLCM)was constructed by fine-tuning Chat-GLM2-6B on the specialized datasets Tianchi TCM,HuangDi,and ShenNong-TCM-Dataset,as well as a TCM lung cancer KG.The KG2TRAG method was applied to enhance the knowledge retrieval,which can convert KG triples into natural language text via ChatGPT-aided linearization,leveraging large language models(LLMs)for context-aware reasoning.For a comprehensive comparison,MedicalGPT,HuatuoGPT,and BenTsao were selected as the baseline models.Performance was evaluated using bilingual evaluation understudy(BLEU),recall-oriented understudy for gisting evaluation(ROUGE),accuracy,and the domain-specific TCM-LCEval metrics,with validation from TCM oncology experts assessing answer accuracy,professionalism,and usability.Results The TCMLCM model achieved the optimal performance across all metrics,including a BLEU score of 32.15%,ROUGE-L of 59.08%,and an accuracy rate of 79.68%.Notably,in the TCM-LCEval assessment specific to the field of TCM,its performance was 3%−12%higher than that of the baseline model.Expert evaluations highlighted superior performance in accuracy and professionalism.Conclusion TCMLCM can provide an innovative solution for TCM lung cancer QA,demonstrating the feasibility of integrating structured KGs with LLMs.This work advances intelligent TCM healthcare tools and lays a foundation for future AI-driven applications in traditional medicine.
基金This work was supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.2022R1F1A1067008)by the Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(No.2019R1A6A1A03032119).
文摘Question-answering(QA)models find answers to a given question.The necessity of automatically finding answers is increasing because it is very important and challenging from the large-scale QA data sets.In this paper,we deal with the QA pair matching approach in QA models,which finds the most relevant question and its recommended answer for a given question.Existing studies for the approach performed on the entire dataset or datasets within a category that the question writer manually specifies.In contrast,we aim to automatically find the category to which the question belongs by employing the text classification model and to find the answer corresponding to the question within the category.Due to the text classification model,we can effectively reduce the search space for finding the answers to a given question.Therefore,the proposed model improves the accuracy of the QA matching model and significantly reduces the model inference time.Furthermore,to improve the performance of finding similar sentences in each category,we present an ensemble embedding model for sentences,improving the performance compared to the individual embedding models.Using real-world QA data sets,we evaluate the performance of the proposed QA matching model.As a result,the accuracy of our final ensemble embedding model based on the text classification model is 81.18%,which outperforms the existing models by 9.81%∼14.16%point.Moreover,in terms of the model inference speed,our model is faster than the existing models by 2.61∼5.07 times due to the effective reduction of search spaces by the text classification model.
文摘Geological strength index(GSI)has been widely used as an input parameter in predicting the strength and deformation properties of rock masses.This study derived a series of equations to satisfy the original GSI lines on the basic GSI chart.Two axes ranging from 0 to 100 were employed for surface conditions of the discontinuities and the structure of rock mass,which are independent of the input parameters.The derived equations can analyze GSI values ranging from 0 to 100 within±5%error.The engineering dimensions(EDs)such as the slope height,tunnel width,and foundation width were used together with representative elementary volume(REV)in jointed rock mass to define scale factor(sf)from 0.2 to 1 in evaluating the rock mass structure including joint pattern.The transformation of GSI into a scaledependent parameter based on engineering scale addresses a crucial requirement in various engineering applications.The improvements proposed in this study were applied to a real slope which was close to the time of failure.The results of stability assessments show that the new proposals have sufficient capability to define rock mass quality considering EDs.
基金Supported by the Inner Mongolia Autonomous Region's Key Research and Achievement Transformation plan (2023YFHH0093).
文摘The quality control chart of fructo-oligosaccharides in milk powder was established to determine whether the detection process and results are in control state.The content of fructo-oligosaccharides in milk powder control samples was determined by ion chromatography,and the quality control chart of fructo-oligosaccharides was established to analyze the controlled state.The results indicate that the median of the quality control chart is 1613.14 mg/100 g,and the standard deviation is 85.57 mg/100 g.The new quality control points were evaluated and analyzed,and the precision changed,but the mean value did not change.Further F test was conducted to determine that the precision did not change significantly,indicating that the test was in a statistical control state,and the detection process,method and results were controlled.
基金the National Science,Research and Innovation Fund(NSRF)King Mongkuts University of Technology North Bangkok under contract no.KMUTNB-FF-68-B-08.
文摘This study aims to examine the explicit solution for calculating the Average Run Length(ARL)on the triple exponentially weighted moving average(TEWMA)control chart applied to autoregressive model(AR(p)),where AR(p)is an autoregressive model of order p,representing a time series with dependencies on its p previous values.Additionally,the study evaluates the accuracy of both explicit and numerical integral equation(NIE)solutions for AR(p)using the TEWMA control chart,focusing on the absolute percentage relative error.The results indicate that the explicit and approximate solutions are in close agreement.Furthermore,the study investigates the performance of exponentially weighted moving average(EWMA)and TEWMA control charts in detecting changes in the process,using the relative mean index(RMI)as a measure.The findings demonstrate that the TEWMA control chart outperforms the EWMA control chart in detecting process changes,especially when the value ofλis sufficiently large.In addition,an analysis using historical data from the SET index between January 2024 and May 2024 and historical data of global annual plastic production,the results of both data sets also emphasize the superior performance of the TEWMA control chart.
基金supported by a grant from the Ministry of Research,Innovation and Digitization,CNCS/CCCDI-UEFISCDI,project number COFUND-CETP-SMART-LEM-1,within PNCDI Ⅳ.
文摘Extracting data from visually rich documents and charts using traditional methods that rely on OCR-based parsing poses multiple challenges,including layout complexity in unstructured formats,limitations in recognizing visual elements,and the correlation between different parts of the documents,as well as domain-specific semantics.Simply extracting text is not sufficient;advanced reasoning capabilities are proving to be essential to analyze content and answer questions accurately.This paper aims to evaluate the ability of the Large Language Models(LLMs)to correctly answer questions about various types of charts,comparing their performance when using images as input versus directly parsing PDF files.To retrieve the images from the PDF,ColPali,a model leveraging state-of-the-art visual languagemodels,is used to identify the relevant page containing the appropriate chart for each question.Google’s Gemini multimodal models were used to answer a set of questions through two approaches:1)processing images derived from PDF documents and 2)directly utilizing the content of the same PDFs.Our findings underscore the limitations of traditional OCR-based approaches in visual document understanding(VrDU)and demonstrate the advantages of multimodal methods in both data extraction and reasoning tasks.Through structured benchmarking of chart question answering(CQA)across input formats,our work contributes to the advancement of chart understanding(CU)and the broader field of multimodal document analysis.Using two diverse and information-rich sources:the World Health Statistics 2024 report by theWorld Health Organisation and the Global Banking Annual Review 2024 by McKinsey&Company,we examine the performance ofmultimodal LLMs across different input modalities,comparing their effectiveness in processing charts as images versus parsing directly from PDF content.These documents were selected due to their multimodal nature,combining dense textual analysis with varied visual representations,thus presenting realistic challenges for vision-language models.This comparison is aimed at assessing how advanced models perform with different input formats and to determine if an image-based approach enhances chart comprehension in terms of accurate data extraction and reasoning capabilities.
文摘We use transaction-level data from the Bitcoin exchange Mt.Gox,including over 1.4 million transactions from more than 45,000 traders,to investigate the role of technical chart patterns in the early Bitcoin market from April 2011 to September 2013.Employing a pattern recognition algorithm,we identify hourly trading signals for five major chart patterns.Buy signals of these patterns are associated with an average increase in abnormal trading volume of more than 53%.Trades executed during buy signal periods yield significantly higher average returns than those made during non-signal periods.Traders who use chart patterns more frequently are more likely to generate right-skewed return distributions,engage in more active trading,and achieve higher average roundtrip returns.Our research suggests that chart pattern trading was a crucial tool for Mt.Gox clients,highlighting the importance of technical heuristics in shaping the dynamics in a less efficient and unregulated market environment.By leveraging a comprehensive transaction dataset from a major cryptocurrency exchange,we provide unique insights into the actual trading behavior of the first Bitcoin adopters.This sets our work apart from previous studies that mainly rely on backtesting technical strategies using publicly available price data.
基金Supported by Korea Health Technology R&D Project through the Korea Health Industry Development Institute(KHIDI),the Ministry of Health and Welfare,Republic of Korea,No.RS-2023-KH142002.
文摘BACKGROUND Restless legs syndrome(RLS)is a neurological disorder characterized by an uncontrollable urge to move the legs during rest.Standard treatments may provide short-term relief but are limited by long-term adverse effects.Herbal medicine containing Paeoniae Radix(HCP)has traditionally been used in Korean medicine to treat RLS symptoms.AIM To investigate the clinical effectiveness and safety of HCP in treating RLS.METHODS This retrospective chart review study included 15 patients with RLS who were treated with HCP between September 2017 and August 2022 at the Kyung Hee University Korean Medicine Hospital.The demographic,disease-related,and treatment-related characteristics,as well as the impressions of each patient,were analyzed.The study also evaluated the changes in the subjective complaints of RLS symptoms,the Korean version of the International Restless Legs Scale(KIRLS)scores before and after treatment,and the adverse events after HCP administration.RESULTS Following HCP administration,14 patients(93.3%)experienced a decrease in the degree of discomfort associated with their symptoms,with an average reduction to 44.7±35.1%compared to their discomfort levels before taking the drug.After treatment,the mean total K-IRLS score decreased significantly from 23.00±9.75 to 13.93±10.9(P=0.005).Among the HCP,a significant reduction in the K-IRLS score was observed in the Gyejibokryeonghwan(26.0±9.7 to 10.0±9.4,P=0.028)and Jakyakgamchotang(23.2±12.6 to 6.6±4.2,P=0.042)groups.Additionally,no adverse events were reported.CONCLUSION HCP improved symptoms in RLS patients.These results suggest that HCP may be an effective and safe alternative for RLS.A randomized controlled trial is required to confirm these findings.
基金financially by the National Research Council of Thailand(NRCT)under Contract No.N42A670894.
文摘This study presents an innovative development of the exponentially weighted moving average(EWMA)control chart,explicitly adapted for the examination of time series data distinguished by seasonal autoregressive moving average behavior—SARMA(1,1)L under exponential white noise.Unlike previous works that rely on simplified models such as AR(1)or assume independence,this research derives for the first time an exact two-sided Average Run Length(ARL)formula for theModified EWMAchart under SARMA(1,1)L conditions,using a mathematically rigorous Fredholm integral approach.The derived formulas are validated against numerical integral equation(NIE)solutions,showing strong agreement and significantly reduced computational burden.Additionally,a performance comparison index(PCI)is introduced to assess the chart’s detection capability.Results demonstrate that the proposed method exhibits superior sensitivity to mean shifts in autocorrelated environments,outperforming existing approaches.The findings offer a new,efficient framework for real-time quality control in complex seasonal processes,with potential applications in environmental monitoring and intelligent manufacturing systems.
