Retrieval Engine Research Articles

A platform-based Natural Language processing-driven strategy for digitalising regulatory compliance processes for the built environment

The digitalisation of the regulatory compliance process has been an active area of research for several decades. However, more recently the level of activities in this area has increased considerably. In the UK, the tragic incident of Grenfell fire in 2017 has been a major catalyst for this as a result of the Hackitt report’s recommendations pointing a lot of the blame on the broken regulatory regime in the country. The Hackitt report emphasises the need to overhaul the building regulations, but the approach to do so remains an open research question. Existing work in this space tends to overlook the processing of actual regulatory documents, or limits their scope to solving a relatively small subtask. This paper presents a new comprehensive platform approach to the digitalisation of the regulatory compliance processing. We present i-ReC (intelligent Regulatory Compliance), a platform approach to digitalisation of regulatory compliance that takes into consideration the enormous diversity of all the stakeholders’ activities. A historical perspective on research in this area is first presented to put things in perspective which identifies the challenges in such an endeavour and identifies the gaps in state-of-the-art. After enumerating all the challenges in implementing a platform-based approach to digitalising the regulatory compliance process, the implementation of some parts of the platform is described. Our research demonstrates that the identification and extraction of all relevant requirements from the corpus of several hundred regulatory documents is a key part of the whole process which underlies the entire process from authoring to eventually compliance checking of designs. Some of the issues that need addressing in this endeavour include ambiguous language, inconsistent use of terms, contradicting requirements and handling multi-word expressions. The implementation of these tools is driven by NLP, ML and Semantic Web technologies. A semantic search engine was developed and validated against other popular and comparable engines with a corpus of 420 (out of about 800) documents used in the UK for compliance checking of building designs. In every search scenario, our search engine performed better on all objective criteria. Limitations of the approach are discussed which includes the challenges around licensing for all the documents in the corpus. Further work includes improving the performance of SPaR.txt (the tool created to identify multi-word expressions) as well as the information retrieval engine by increasing the dataset and providing the model with examples from more diverse formats of regulations. There is also a need to develop and align strategies to collect a comprehensive set of domain vocabularies to be combined in a Knowledge Graph.

Interactive multimodal video search: an extended post-evaluation for the VBS 2022 competition

CLIP-based text-to-image retrieval has proven to be very effective at the interactive video retrieval competition Video Browser Showdown 2022, where all three top-scoring teams had implemented a variant of a CLIP model in their system. Since the performance of these three systems was quite close, this post-evaluation was designed to get better insights on the differences of the systems and compare the CLIP-based text-query retrieval engines by introducing slight modifications to the original competition settings. An extended analysis of the overall results and the retrieval performance of all systems’ functionalities shows that a strong text retrieval model certainly helps, but has to be coupled with extensive browsing capabilities and other query-modalities to consistently solve known-item-search tasks in a large-scale video database.

Multi-level and cross-domain search engine based on graph convolutional neural network

Abstract With the advent of the information age, there is a growing demand for multi-level cross-domain information retrieval by users. However, current technological solutions have not yet overcome the dilemma of automatic adjustment of search scope in cross-domain retrieval. This article proposes a multi-level cross-domain retrieval engine based on graph convolutional neural networks (GCNs). GCN can effectively capture the complex semantic structures in documents, excavate the potential dynamic context of documents by learning the relationships between nodes in the graph structure, adapt to cross-domain retrieval, extract semantic information from different levels and automatically adjust the search scope. To verify the effectiveness of the method, experiments were conducted on the NQ320k dataset and compared with SEAL, BM25, Transformer and Bert models. The results show that the cross-domain retrieval engine based on GCN has significant advantages over other methods, with Recall@100 reaching 97.44%, MRR@100 reaching 76.12% and R-Precision reaching 78.84%. In the future, the performance and applicability of the retrieval engine can be further improved by optimizing the model structure and expanding the training dataset.

Comparative study on generation of attosecond pulse train and phase information reconstruction

<sec>Attosecond pulses provide higher measurement precision for analyzing ultrafast dynamics in atoms, molecules, and electrons, laying the foundation for studying electronic motion in atomic and molecular systems. The most important method currently is to generate attosecond pulse trains and isolated attosecond pulses through the interaction of femtosecond lasers with gases. The temporal information of attosecond pulses and the dynamic information of electrons can be extracted from spectrograms by using attosecond streak camera or the reconstruction of attosecond beating by interference of two-photon transitions (RABBITT) method based on two-photon transition interference. Although the differences in phase among different high-order harmonics can be directly extracted from the oscillation frequencies of sidebands, the iterative algorithm of attosecond streak camera can provide complete phase information of attosecond pulse trains to better support the study of electron dynamics in atoms.</sec> <sec> <b>Research purpose</b> This work is dedicated to the investigation of the generation, measurement, and characterization of attosecond pulse train (APT), which are essential for probing ultrafast dynamics in atomic, molecular, and electronic systems. The focus is on the generation of APTs through interactions between femtosecond lasers and gases, as well as the extraction of temporal and dynamic information from these pulses by using advanced spectroscopic techniques such as the RABITT method.</sec> <sec> <b>Methods</b> The experimental approach involves the use of a homebuilt femtosecond titanium sapphire regenerative amplifier to produce high-order harmonics, leading to the generation of APTs. The setup includes the homebuilt titanium sapphire chirped pulse amplifier and a collinear attosecond pulse generation and measurement beamline, which are used to conduct RABITT experiments. The process requires the interaction of femtosecond lasers with gas targets to generate high-energy photons in the extreme ultraviolet and soft X-ray spectral ranges. By optimizing the phase-matching conditions within the gas target, strong high-order harmonic signals are observed on an XUV spectrometer. The temporal information of the attosecond pulses is indirectly measured through the photoelectron spectrum produced by the interaction of attosecond pulses with femtosecond lasers. The research also employs the FROG-CRAB algorithm and the extended phase retrieval and iterative engine (ePIE) algorithm for temporally reconstructing APTs and attempts to use a genetic algorithm to extract phase information.</sec> <sec> <b>Results</b> The study yields three sets of RABITT spectrograms, which are analyzed by using the RABITT sideband phase method to directly reconstruct APTs. Fourier transform analysis is used to extract phase differences between sidebands, offering insights into the phase differences between corresponding high-order harmonics. This method, however, provides an estimation of the phase in the center of each harmonic order, which does not fully represent the actual pulse shape. The FROG-CRAB algorithm and ePIE algorithm successfully reconstructs the attosecond pulse trains from the RABITT spectrograms, revealing similar temporal pulse train morphologies. In contrast, the genetic algorithm, despite its potential for high constraint optimization, does not yield satisfactory results, possibly due to the sensitivity of the algorithm to discrepancies between theoretical simulations and experimental data.</sec> <sec> <b>Conclusions</b> The research concludes that achieving ideal inversion results for APTs necessitates small time delay steps and a wide scanning range in the experimental data collection process to ensure a rich dataset for inversion. The FROG-CRAB algorithm and ePIE algorithm demonstrate their effective performance in reconstructing APTs, with ePIE showing higher computational efficiency. The genetic algorithm, while offering a high degree of constraint, faces challenges and requires to be further refined. The study underscores the importance of the signal-to-noise ratio in experimental data for the accuracy of inversion results. This work provides significant guidance for future measuring electron dynamics and explaining their evolution patterns, contributing valuable experimental methods and data analysis techniques to the field of attosecond science.</sec>

Is it the new Google: Impact of ChatGPT on Students' Information Search Habits

The traditional subject of information search and retrieval (IR) paradigm shifted to an entirely new era since artificial intelligence (AI) techniques were introduced into the field. Browser-based IR solutions powered by AI for personalised recommendations-based information retrieval, such as the Google search engine, were one of the early examples. The IR field has advanced to its next level with the newest conversational applications based on large language model (LLM) techniques. It is becoming clear that Generative Pretrained Transformer (GPT) applications such as ChatGPT will significantly impact information retrieval behaviour in the education sector. Though this application has become widespread in acclaim, no previous study has shown its impact on information seeking and retrieval. However, based on the observation of the fast penetration of this technology and the growth of public interest, a pre-assumption was built that it is essential to investigate if students may also be showing a similar interest in this new tool. Hence, this study is set up to systematically and empirically explore how ChatGPT influences the IR behaviour of students in HEIs. A survey approach is utilised to collect the perceived IR behaviour through a questionnaire administered to 60 students in HEIs. The findings reveal that the tool is already widely known among HEI students. They also perceived the use of the tool in the context of information retrieval and proclaimed its usefulness, acknowledging its efficiency (reduced time) in finding information. Furthermore, the technology has considerably affected the typical use of other conventional information retrieval and search engine tools. On the contrary, 10% of the respondents are less likely to use ChatGPT during information seeking for various reasons, from credibility and relevance to technology infrastructure issues such as connectivity. Although a deeper analysis is required to establish a general conclusion on how and in which ways GPT-based models will override contemporary IR practices, the study outcome provides evidence for a possible behavioural change among HEI students in their IR habits in the future.

Similar Image Retrieval using Autoencoder. I. Automatic Morphology Classification of Galaxies

We present the construction of an image similarity retrieval engine for the morphological classification of galaxies using the Convolutional AutoEncoder (CAE). The CAE is trained on 90,370 preprocessed Sloan Digital Sky Survey galaxy images listed in the Galaxy Zoo 2 (GZ2) catalog. The visually similar output images returned by the trained CAE suggest that the encoder efficiently compresses input images into latent features, which are then used to calculate similarity parameters. Our Tool for Searching a similar Galaxy Image based on a Convolutional Autoencoder using Similarity (TSGICAS) leverages this similarity parameter to classify galaxies’ morphological types, enabling the identification of a wider range of classes with high accuracy compared to traditional supervised ML techniques. This approach streamlines the researcher’s work by allowing quick prioritization of the most relevant images from the latent feature database. We investigate the accuracy of our automatic morphological classifications using three galaxy catalogs: GZ2, Extraction de Formes Idéalisées de Galaxies en Imagerie (EFIGI), and Nair & Abraham (NA10). The correlation coefficients between the morphological types of input and retrieved galaxy images were found to be 0.735, 0.811, and 0.815 for GZ2, EFIGI, and NA10 catalogs, respectively. Despite differences in morphology tags between input and retrieved galaxy images, visual inspection showed that the two galaxies were very similar, highlighting TSGICAS’s superior performance in image similarity search. We propose that morphological classifications of galaxies using TSGICAS are fast and efficient, making it a valuable tool for detailed galaxy morphological classifications in other imaging surveys.

Higher Vocational Students’ Innovation and Entrepreneurship Ability Demand Prediction

The new era background and new social background put forward higher goals for the cultivation of innovative talents. The accurate prediction of the demand for innovation and entrepreneurship vocational ability of higher vocational students provides an important reference for higher vocational educators to formulate refined innovation and entrepreneurship education management strategies, more balanced allocation of limited innovation and entrepreneurship education resources, and make more scientific innovation and entrepreneurship education decisions. This paper studies the prediction of innovation and entrepreneurship vocational ability demand of higher vocational students under the background of digital informatization and theoretically discusses the demand for innovation and entrepreneurship vocational ability of higher vocational students and its connotation. It constructs the demand prediction framework of innovation and entrepreneurship vocational ability of higher vocational students for data identified on the Knowledge Retrieval Engine or High-quality Knowledge Q&A and Sharing Community, and develops the collection and use methods of students' innovation and entrepreneurship resource retrieval data. It makes the feature selection of the keyword sequence and the corresponding time lag sequence selected by the higher vocational students' innovation and entrepreneurship resource retrieval and builds the prediction model of the vocational ability demand for innovation and entrepreneurship of higher vocational students. Experimental results verify the effectiveness of the proposed model.

LexID: The Metadata and Semantic Knowledge Graph Construction of Indonesian Legal Document

The Legal Fiction principle stipulates that the government needs to ensure the public availability of all of their legal documents. Unfortunately, the text-based search services they provide cannot return satisfactory answers in retrieval scenarios requiring proper representation of relationships between various legal documents. A key problem here is the lack of explicit representation of such relationships behind the employed retrieval engines. We aim to address this problem by proposing LexID knowledge graph (KG) that provides an explicit knowledge representation for Indonesian legal domain usable for such retrieval purposes. The KG contains both legal metadata information and semantic content of the legal clauses of the legal document's articles, modeled using formal vocabulary from the LexID ontology also presented in this paper. The KG is constructed from thousands of Indonesian legal documents. Since the procedure of writing a legal document regulated by the government is clear and detailed, we use a rule-based approach to construct our KG. At the end, we describe several use cases of the KG to address different retrieval needs. In Addition, we evaluated the quality of our KG by measuring its ability to answer questions and got that LexID can answer questions with the macro average F1 score is about 0.91.

Information Retrieval Based Legal Search System

Calculating the similarity between two legal documents to find similar legal judgments is an important challenge in legal information. Efficiently computing this similarity by expanding widely used information retrieval and search engine techniques has practical applications in a number of tasks, like locating pertinent prior cases for a specific case document. Programmed data recovery frameworks or reports are the main parts of today’s selected emotional support networks or web indexes to reduce data overload. Investigating methodologies to work on the presentation of report recovery frameworks and web search tools is a working area of research. Various methods have been pro- posed in this research paper to explore ways to search the common law system for cases with a similar outcome. Building a legal decision support system is intended to increase efficiency by assisting stakeholders—including judges and attorneys—in finding related rulings promptly. In order to prepare arguments, a lawyer typically has to review earlier decisions that are comparable to (or pertinent to) the current case. The attorney examines the judgement database to discover similar judgements. Legal rulings are complex in nature and refer to other judgments. For this, proper techniques are needed for quality analysis of judgments and correct deductions from them. A proper analysis of several types of similarity measures, such as all-term-based similarity methods, legal terms, co-citations, and bibliographic links, performed to look for comparable conclusions. According to experimental findings, the law term similarity approach outperforms all term cosine similarity methods. The out- comes also demonstrate that the co-citation approach performs worse than the bibliographic linkage similarity method and improves performance over the co-citation approach. After proper analysis of various methods in this field, proper comparison can be made between documents and similar legal documents can also be easily searched based on their similarity pattern and can be used to make meaningful deductions.

Retracted] Multisensor Cross‐Media Data Mining Method Assisted by Expert System

The information generated by the multifaceted and multilevel processing of the data of multiple sensors is more meaningful than the information obtained by a single sensor and provides accurate information and decision‐making basis for various application systems. How to apply data mining theory to the multisensor cross‐media field has become a research hotspot. Through the analysis of multisensor cross‐media data, it is of great significance to dig out the important rules, information, or knowledge hidden in it and use it for cross‐media retrieval engine. The rule acquisition is the “bottleneck” problem of the expert system. This paper adopts the data mining method based on the rough set to acquire the rules and improves the basic algorithm of attribute reduction. Using the attribute reduction algorithm and the heuristic value reduction algorithm, the calculation is simplified and the reduction efficiency is improved. In the presentation, according to the characteristics of cross‐media and the application requirements of expert systems, this paper takes the case representation based on features as the basis and classifies cases according to feature attributes. In case retrieval, according to the hierarchical structure of case features, the entire case database is organized into a multilevel hierarchical index structure. In this paper, a cross‐media retrieval engine is constructed from the perspective of classifier design, and the Euclidean distance is used as the similarity matching model of image content. The mutual retrieval of images and audios preliminarily forms the design process of retrieval from one media type to another and establishes a corresponding cross‐media index. The experimental results show that the algorithm has better processing effect and higher accuracy than other algorithms. Different k‐nearest neighbor values were selected in the experiment, and it reached about 96% in the test environment of libsvm toolbox, which is better than the processing results of LE and LLE algorithms.

Eliminating local convergences in FROG retrieval algorithms

The complicated electric field structure of ultrashort pulses as characterized by frequency-resolved optical gating (FROG) algorithms often results in local convergence, which does not accurately represent the actual field pulse. We present an efficient and universal test procedure applicable to any FROG retrieval algorithm that allows recognition of the erroneous local convergence. The 100% efficacy of the procedure is demonstrated using Line-Search FROG algorithm, and comparison is given with the performance of a standard extended ptychographic iterative FROG trace retrieval engine.

Methodology for development of an expert system to derive knowledge from existing nature-based solutions experiences

Learning from past experiences is essential for the adoption of Nature-Based Solutions (NBS). There is a growing number of knowledge repositories sharing the experience of NBS projects implemented worldwide. These repositories provide access to a large amount of information, however, acquiring knowledge from them remains a challenge. This paper outlines the technical details of the NBS Case-Based System (NBS-CBS), an expert system that facilitates knowledge acquisition from an NBS case repository. The NBS-CBS is a hybrid system integrating a black-box Artificial Neural Network (ANN) with a white-box Case-Based Reasoning model. The system involves:•a repository that stores the information of past NBS projects, and an input collection component, guiding the collection and encoding of the user's inputs;•a classifier that predicts solutions (i.e., generates a hypothesis), based on user input (target case), drawing on a pre-trained ANN model to guide the case retrieval, and a case retrieval engine that identifies cases similar to the target case;•a case adaption and retainment process in which the user assesses the provided recommendations and retains the solved problem as a new case in the repository.

PM-Search: State-of-the-art Evidence Retriever for Precision Medicine: Algorithm Development and Validation.

Under the paradigm of Precision Medicine (PM), patients with the same disease can receive different personalized therapies according to their clinical and genetic features. These therapies are determined by the totality of all available clinical evidence, including results from case reports, clinical trials, and systematic reviews. However, it is increasingly difficult for physicians to find such evidence from scientific publications, whose size is growing at an unprecedented pace. In this work, we propose the PM-Search system to facilitate the retrieval of clinical literature that contains critical evidence for or against giving specific therapies to certain cancer patients. The PM-Search system combines a Baseline Retriever that selects document candidates at a large scale and an Evidence Re-ranker that finely reorders the candidates based on their evidence quality. The Baseline Retriever uses query expansion and keyword matching with the Elasticsearch retrieval engine, and the Evidence Re-ranker fits pre-trained language models to expert annotations that are derived from an active learning strategy. The PM-Search system achieves the best performance in the retrieval of high-quality clinical evidence at the TREC PM Track 2020, outperforming the second-ranking systems by large margins (0.4780 vs. 0.4238 for standard NDCG@30 and 0.4519 vs. 0.4193 for exponential NDCG@30). We present PM-Search, a state-of-the-art search engine to assist the practicing of evidence-based PM. PM-Search uses a novel BioBERT-based active learning strategy that models evidence quality and improves the model performance. Our analyses show that evidence quality is a distinct aspect from general relevance, and specific modeling of evidence quality beyond general relevance is required for a PM search engine.

Can BERT Dig It? Named Entity Recognition for Information Retrieval in the Archaeology Domain

The amount of archaeological literature is growing rapidly. Until recently, these data were only accessible through metadata search. We implemented a text retrieval engine for a large archaeological text collection (~658 million words). In archaeological IR, domain-specific entities such as locations, time periods and artefacts play a central role. This motivated the development of a named entity recognition (NER) model to annotate the full collection with archaeological named entities. In this article, we present ArcheoBERTje, a BERT (Bidirectional Encoder Representations from Transformers) model pre-trained on Dutch archaeological texts. We compare the model’s quality and output on an NER task to a generic multilingual model and a generic Dutch model. We also investigate ensemble methods for combining multiple BERT models, and combining the best BERT model with a domain thesaurus using conditional random fields. We find that ArcheoBERTje outperforms both the multilingual and Dutch model significantly with a smaller standard deviation between runs, reaching an average F1 score of 0.735. The model also outperforms ensemble methods combining the three models. Combining ArcheoBERTje predictions and explicit domain knowledge from the thesaurus did not increase the F1 score. We quantitatively and qualitatively analyse the differences between the vocabulary and output of the BERT models on the full collection and provide some valuable insights in the effect of fine-tuning for specific domains. Our results indicate that for a highly specific text domain such as archaeology, further pre-training on domain-specific data increases the model’s quality on NER by a much larger margin than shown for other domains in the literature, and that domain-specific pre-training makes the addition of domain knowledge from a thesaurus unnecessary.

Intuitively Searching for the Rare Colors from Digital Artwork Collections by Text Description: A Case Demonstration of Japanese Ukiyo-e Print Retrieval

In recent years, artworks have been increasingly digitized and built into databases, and such databases have become convenient tools for researchers. Researchers who retrieve artwork are not only researchers of humanities, but also researchers of materials science, physics, art, and so on. It may be difficult for researchers of various fields whose studies focus on the colors of artwork to find the required records in existing databases, that are color-based and only queried by the metadata. Besides, although some image retrieval engines can be used to retrieve artwork by text description, the existing image retrieval systems mainly retrieve the main colors of the images, and rare cases of color use are difficult to find. This makes it difficult for many researchers who focus on toning, colors, or pigments to use search engines for their own needs. To solve the two problems, we propose a cross-modal multi-task fine-tuning method based on CLIP (Contrastive Language-Image Pre-Training), which uses the human sensory characteristics of colors contained in the language space and the geometric characteristics of the sketches of a given artwork in order to gain better representations of that artwork piece. The experimental results show that the proposed retrieval framework is efficient for intuitively searching for rare colors, and that a small amount of data can improve the correspondence between text descriptions and color information.

Content-based medical image retrieval system for lung diseases using deep CNNs.

Content-based image retrieval (CBIR) systems are designed to retrieve images that are relevant, based on detailed analysis of latent image characteristics, thus eliminating the dependency of natural language tags, text descriptions, or keywords associated with the images. A CBIR system maintains high-level image visuals in the form of feature vectors, which the retrieval engine leverages for similarity-based matching and ranking for a given query image. In this paper, a CBIR system is proposed for the retrieval of medical images (CBMIR) for enabling the early detection and classification of lung diseases based on lung X-ray images. The proposed CBMIR system is built on the predictive power of deep neural models for the identification and classification of disease-specific features using transfer learning based models trained on standard COVID-19 Chest X-ray image datasets. Experimental evaluation on the standard dataset revealed that the proposed approach achieved an improvement of 49.71% in terms of precision, averaging across various distance metrics. Also, an improvement of 26.55% was observed in the area under precision-recall curve (AUPRC) values across all subclasses.

Structural Information Retrieval in XML Documents: A Graph-based Approach

Although retrieval engines are becoming more and more functional and efficient, they still have the drawback of not being able to locate the relevant documentary granularity, which results in ignoring the structural aspect. In the context of XML document, Information Retrieval Systems allow to return the user’s documentary granules. Several studies have used graphs to represent XML documents. However, in the scope of this research, the semi-structured document’s structure and that of a user’s query can be seen as arborescences composed of a hierarchy of nested elements. By using graph theory, by calculating the structural proximity and especially the intersection between these two arborescences. The article presents a model for structural information retrieval based on graphs. A collection of multimedia documents are randomly extracted from INEX (Initiative for the Evaluation of XML Retrieval) 2010 to validate the approach. The first results shows the interest of such an approach.

Quality assessment of conventional and traditional oriental medicine clinical practice guidelines for knee osteoarthritis using AGREE II instrument.

Introduction:Knee osteoarthritis is a degenerative disease and its prevalence tends to increase. Clinical practice guidelines (CPGs) are evidence-based recommendations for treatment that help policymakers, practitioners, and patients make more appropriate and efficient decisions during the course of management. This study aimed to evaluate the quality of knee osteoarthritis CPGs using the Appraisal of Guidelines for Research and Evaluation (AGREE II) instrument.Method:The retrieval engines and websites were utilized from January 2010 to December 2020. The search words were “Clinical practice guideline” OR “Critical practice guideline” OR “guideline∗” AND “Osteoarthritis.” The quality of the CPGs was independently examined by four appraisers using the AGREE II instrument. Consequently, the selected CPGs were graded as Classes A, B, and C according to the level of recommendation.Result:In this study, 13 CPGs for knee osteoarthritis were selected and evaluated qualitatively using the AGREE II instrument. The overall quality percentage score was as follows: clarity of presentation, 72.6%, scope and purpose, 62.6%, rigor of development, 54.2%, stakeholder investment, 50.5%, editorial independence, 46.5%, applicability, 22.5%.Conclusion:Auxiliary materials for the treatment process of knee OA should be supplemented in future revised versions for quality improvement of knee OA CPGs. Also, more evidence should be accumulated to support the recommendation of traditional oriental medical treatments in the clinical field. From the perspective of integrative medicine, along with conventional pharmacological treatment, exercise, weight loss, and acupuncture can be combined together in clinical situations.

AMFormulaS: an intelligent retrieval system for traditional Chinese medicine formulas

BackgroundFormula is an important means of traditional Chinese medicine (TCM) to treat diseases and has great research significance. There are many formula databases, but accessing rich information efficiently is difficult due to the small-scale data and lack of intelligent search engine.MethodsWe selected 38,000 formulas from a semi-structured database, and then segmented text, extracted information, and standardized terms. After that, we constructed a structured formula database based on ontology and an intelligent retrieval engine by calculating the weight of decoction pieces of formulas.ResultsThe intelligent retrieval system named AMFormulaS (means Ancient and Modern Formula system) was constructed based on the structured database, ontology, and intelligent retrieval engine, so the retrieval and statistical analysis of formulas and decoction pieces were realized.ConclusionsAMFormulaS is a large-scale intelligent retrieval system which includes a mass of formula data, efficient information extraction system and search engine. AMFormulaS could provide users with efficient retrieval and comprehensive data support. At the same time, the statistical analysis of the system can enlighten scientific research ideas and support patent review as well as new drug research and development.

Smart integration of sensors, computer vision and knowledge representation for intelligent monitoring and verbal human-computer interaction.

The details presented in this article revolve around a sophisticated monitoring framework equipped with knowledge representation and computer vision capabilities, that aims to provide innovative solutions and support services in the healthcare sector, with a focus on clinical and non-clinical rehabilitation and care environments for people with mobility problems. In contemporary pervasive systems most modern virtual agents have specific reactions when interacting with humans and usually lack extended dialogue and cognitive competences. The presented tool aims to provide natural human-computer multi-modal interaction via exploitation of state-of-the-art technologies in computer vision, speech recognition and synthesis, knowledge representation, sensor data analysis, and by leveraging prior clinical knowledge and patient history through an intelligent, ontology-driven, dialogue manager with reasoning capabilities, which can also access a web search and retrieval engine module. The framework’s main contribution lies in its versatility to combine different technologies, while its inherent capability to monitor patient behaviour allows doctors and caregivers to spend less time collecting patient-related information and focus on healthcare. Moreover, by capitalising on voice, sensor and camera data, it may bolster patients’ confidence levels and encourage them to naturally interact with the virtual agent, drastically improving their moral during a recuperation process.