Strong Semantics Research Articles

ProFPN: Progressive feature pyramid network with soft proposal assignment for object detection

Benefitting from the development of pyramidal feature learning, current state-of-the-art multi-scale detection paradigm has become proficient in detecting objects of varying scales. However, feature pyramid network (FPN), in spite of constructing multi-scale features with strong semantics, still suffers from limited performance caused by insufficient detail exploitation, information loss, limited receptive fields and hard proposal assignment, which can be mainly categorized into semantic level and instance level. To address these limitations, this paper analyzes the structural components that inhibit multi-scale feature representation and then presents a multi-stage progressive FPN (ProFPN) along with a novel RoI feature representation method called soft proposal assignment. In the semantic level, the bottom-up interaction module is first proposed to address to insufficient exploitation of high resolution features. In the bottom-up interaction module, global context attention blocks are utilized to interact adjacent-level features with detail information in a bottom-up progressive manner. After that, the top-down transfer module is designed to mitigate semantic information loss of high-level features. In the top-down transfer module, multi-branch asymmetric dilated blocks are adopted in a top-down progressive manner, which expands receptive fields to capture more object poses. In the instance level, to overcome the hard assignment of object proposals, a nonparametric strategy named soft proposal assignment is proposed to leverage the scale of each object proposal to generate dynamic weights for RoI features from adjacent levels. Comprehensive experiments conducted on MS COCO dataset demonstrate the superiority of ProFPN. By adding negligible extra FLOPs, the proposed ProFPN outperforms most pyramid-based methods. Moreover, due to the design of inherited feature utilization in ProFPN, transformer-based detectors have witnessed a substantial increase in detecting small objects while simultaneously achieving significant reductions in FLOPs. The source code of the proposed method is available at https://github.com/GingerCohle/ProFPN.

Multi-Scale Feature Pyramid for Detection of Red Lesions in Fundus Images

Diabetic retinopathy (DR) is increasing rapidly around the world, but there is a shortage of experienced ophthalmologists. Therefore, computer-based diagnosis of the fundus images is essential to screening of referable DR. Automated detection of red lesions is very important for screening of DR. This paper deals with a novel method for automatic detection of red lesion. The main contribution is developing a deep learning based detection framework to handle severe class imbalance and imbalance in sizes of red lesions. The multi-scale features are extracted using the feature pyramid network. A pyramid of features is generated with strong semantics. The proposed network is end-to-end trainable in image level with several scales and works for a wide range of red lesions with acceptable performance. Sensitivity of the proposed method is 0.76 with six false-positive per image on test set of publicly available DIARECTDB1 database and outperforms state-of-the-art approaches. A potential benefit with deep learning based detection framework could be used in screening programs of referable DR.

SliceSamp: A Promising Downsampling Alternative for Retaining Information in a Neural Network

Downsampling, which aims to improve computational efficiency by reducing the spatial resolution of feature maps, is a critical operation in neural networks. Many downsampling methods have been proposed to address the challenge of retaining feature map information. However, some detailed information is still lost, even though these methods can extract features with stronger semantics. In this paper, we propose a novel downsampling method which combines feature slicing and depthwise separable convolution for information-retaining downsampling. It slices the input feature map into multiple non-overlapping sub-feature maps by using indexes with a stride of two in the spatial dimension and applies depthwise separable convolution on each slice to extract feature information. To demonstrate the effectiveness of SliceSamp, we compare it with classical downsampling methods on image classification, object detection, and semantic segmentation tasks using several benchmark datasets, including ImageNet-1K, COCO, VOC, and ADE20K. Extensive experiments demonstrate that SliceSamp outperforms classical downsampling methods with consistent improvements in various computer vision tasks. The proposed SliceSamp shows advanced model performance with lower computational costs and memory requirements. By replacing the downsampling layers in different network architectures (including ResNet (Residual Network), YOLOv5, and Swin Transformer), SliceSamp brings different degrees of performance gains (+0.54~3.64%) compared to these baseline models. Additionally, SliceUpsamp enables high-resolution feature reconstruction and alignment during upsampling. SliceSamp and SliceUpsamp can be plug-and-play-integrated into existing neural network architectures. As a promising downsampling alternative to traditional methods, SliceSamp can also provide a reference for designing lightweight and high-performance model architectures in the future.

A new method for safety helmet detection based on convolutional neural network.

Considering practical issues such as cost control of hardware facilities in engineering projects, it is a challenge to design a robust safety helmet detection method, which can be implemented on mobile or embedded devices with limited computing power. This paper presents an approach to optimize the BottleneckCSP structure in the YOLOv5 backbone network, which can greatly reduce the complexity of the model without changing the size of the network input and output. To eliminate the information loss caused by upsampling and enhance the semantic information of the feature map on the reverse path, this paper designs an upsampling feature enhancement module. Besides, To avoid the negative impact of redundant information generated by feature fusion on the detection results, this paper introduces a self-attention mechanism. That is, using the designed channel attention module and location attention module, adjacent shallow feature maps and upsampled feature maps are adaptively fused to generate new feature maps with strong semantics and precise location information. Compared with the existing methods with the fastest inference speed, under the same compute capability, the proposed method not only has faster inference speed, the FPS can reach 416, but also has better performance with mAP of 94.2%.

THE SEMANTIC GRAMMATICAL DIMENSIONS OF AL-RUMANI IN HIS EXPLANATION OF SIBAWAYH'S BOOK

The grammatical significance is an extension of the sentence system, arrangement and syntactic movements ([i]), or is the relationships between the words of one sentence, and the significance divided by Ibn Jinni in his characteristics on three ranks, according to the force, "the strongest semantics and then followed by industrial and then followed by moral ... Do you not see that he has risen, and the indication of his word on its source, and the indication of its construction on its time, and the indication of its meaning on its subject, for these are three indications of its word, its form and its meaning."[ii]
 
 Semantic[i] development between the language of pre-Islamic poetry and the language of the Holy Qur'an: 276.
 Characteristics[ii]: 3/100.

Influential Global and Local Contexts Guided Trace Representation for Fault Localization

Trace data is critical for fault localization (FL) to analyze suspicious statements potentially responsible for a failure. However, existing trace representation meets its bottleneck mainly in two aspects: (1) the trace information of a statement is restricted to a local context (i.e., a test case) without the consideration of a global context (i.e., all test cases of a test suite); (2) it just uses the ‘occurrence’ for representation without strong FL semantics. Thus, we propose UNITE : an infl U ential co N text-Gu I ded T race r E presentation, representing the trace from both global and local contexts with influential semantics for FL. UNITE embodies and implements two key ideas: (1) UNITE leverages the widely used weighting capability from local and global contexts of information retrieval to reflect how important a statement (a word) is to a test case (a document) in all test cases of a test suite (a collection), where a test case (a document) and all test cases of a test suite (a collection) represent local and global contexts respectively; (2) UNITE further elaborates the trace representation from ‘occurrence’ (weak semantics) to ‘influence’ (strong semantics) by combing program dependencies. The large-scale experiments on 12 FL techniques and 20 programs show that UNITE significantly improves FL effectiveness.

SFAF-MA: Spatial Feature Aggregation and Fusion With Modality Adaptation for RGB-Thermal Semantic Segmentation

The fusion of RGB and thermal images has profound implications for the semantic segmentation of challenging urban scenes, such as those with poor illumination. Nevertheless, existing RGB-T fusion networks pay less attention to modality differences; i.e., RGB and thermal images are commonly fused with fixed weights. In addition, spatial context details are lost during regular extraction operations, inevitably leading to imprecise object segmentation. To improve the segmentation accuracy, a novel network named spatial feature aggregation and fusion with modality adaptation (SFAF-MA) is proposed in this paper. The modality difference adaptive fusion (MDAF) module is introduced to adaptively fuse RGB and thermal images with corresponding weights generated from an attention mechanism. In addition, the spatial semantic fusion (SSF) module is designed to tap into more information by capturing multiscale perceptive fields with dilated convolutions of different rates, and aggregate shallower-level features with rich visual information and deeper-level features with strong semantics. Compared with existing methods on the public MFNet dataset and PST900 dataset, the proposed network significantly improves the segmentation effectiveness. The code is available at https://github.com/hexunjie/SFAF-MA.

Automated optical inspection of FAST’s reflector surface using drones and computer vision

The Five-hundred-meter Aperture Spherical radio Telescope (FAST) is the world ’ s largest single-dish radio telescope. Its large reflecting surface achieves unprecedented sensitivity but is prone to damage, such as dents and holes, caused by naturally-occurring falling objects. Hence, the timely and accurate detection of surface defects is crucial for FAST’s stable operation. Conventional manual inspection involves human inspectors climbing up and examining the large surface visually, a time-consuming and potentially unreliable process. To accelerate the inspection process and increase its accuracy, this work makes the first step towards automating the inspection of FAST by integrating deep-learning techniques with drone technology. First, a drone flies over the surface along a predetermined route. Since surface defects significantly vary in scale and show high inter-class similarity, directly applying existing deep detectors to detect defects on the drone imagery is highly prone to missing and misidentifying defects. As a remedy, we introduce cross-fusion, a dedicated plug-in operation for deep detectors that enables the adaptive fusion of multi-level features in a point-wise selective fashion, depending on local defect patterns. Consequently, strong semantics and fine-grained details are dynamically fused at different positions to support the accurate detection of defects of various scales and types. Our AI-powered drone-based automated inspection is time-efficient, reliable, and has good accessibility, which guarantees the long-term and stable operation of FAST.

Research on intelligent recognition method for self‐blast state of glass insulator based on mixed data augmentation

AbstractAutomatically and accurately detecting the self‐blast state of glass insulators is of great significance to operation and maintenance of transmission lines. To solve the shortcomings of the existing open‐loop cognitive models to detect the self‐blast state of glass insulators, this study explores a mixed data augmentation‐based intelligent recognition method to detect the self‐blast state of the glass insulator, by imitating the human cognitive mode. Firstly, generative adversarial network is utilised to obtain the high‐quality generative self‐blast samples of the glass insulator, and the non‐generative data augmentation techniques is used to obtain rich sample features. Secondly, considering the characteristics of aerial images such as large scale variations, variable shooting angles and complex backgrounds, feature maps with strong semantics and adaptive multi‐scale fusion are extracted using the feature pyramid network with adaptive hierarchy and the multi‐deformable convolutional network. Then, the extracted feature maps are transmitted to a two‐dimensional stochastic configuration network that can adaptively generate hidden nodes and basis functions so as to develop the self‐blast state classification criteria with universal approximation capability. Thirdly, based on the generalised error and entropy theory, the semantic error entropy evaluation indices of recognition results are defined to evaluate in real time, the credibility of the uncertain recognition results for the self‐blast state of the glass insulator. Then, based on transfer learning and the established self‐optimising feedback mechanism for feature pyramid network, the self‐optimising adjustment and reconstruction of the feature map space with strong semantics and multi‐scale fusion and its classification criteria are realised. Finally, the stacking method is applied to integrate the recognition results of the feature pyramid network with adaptive hierarchy and multi‐channel deformable convolutional networks to improve the robustness of the recognition model. Results of experimental comparison with other machine learning and deep learning methods verify the feasibility and effectiveness of the proposed method.

An improved SSD method for infrared target detection based on convolutional neural network

Target detection is the basis for automatic target recognition system of infrared imaging guidance to complete subsequent tasks such as recognition and tracking. Existing systems have not the autonomous learning ability of target feature, and it will be powerless once the task environment exceeds the pre-planned condition. The single-stage target detection based on deep learning has the ability of autonomous learning and high computational efficiency, which is an effective way to solve the problem of infrared imaging guidance target detection in complex environment. SSD (Single Shot MultiBox Detector) is a classical single-stage detection model, however, the convolution layer with strong semantic information in SSD has low resolution, which is not conducive to small target detection. In addition, the location loss of SSD does not consider the impact of target scale change. Therefore, this paper puts forward two improvement ideas in view of SSD: (1) Starting from FPN (Feature Pyramid Network), feature channel’s importance is distinguished through efficient channel attention mechanism, the contribution of each feature layer to the fusion output is described based on the learnable weight, and the feature weighted fusion of bidirectional multi-scale is realized between the feature layer which has low resolution and strong semantics and the feature layer which has high resolution and weak semantics. (2) Starting from IoU (Intersection over Union) and considering non overlapping parts and geometric relationship between the predicted box and the ground-truth box, the location loss of SSD that remains invariable to the target scale change is constructed to improve the sensitivity of the detection model to the locating error of small target. The experimental results show that, for 300 × 300 input, the presented method achieves 84.7% mAP (mean Average Precision) on VOC2007 test and for 512 × 512 input, it reaches 86.6%. On the self-built infrared aircraft data set, the proposed method achieves 81.1% mAP and can detect more small targets. Without affecting detection speed, the presented method on experimental results outperforms some comparable state-of-the-art models such as YOLOv3 (You Only Look Once), DSSD (Deconvolutional Single Shot Multibox Detector), RSSD (Rainbow Single Shot Multibox Detector) and FSSD (Fusion Single Shot Multibox Detector).

The Internet of Things Term Feature Extraction Method and Its Application in English-Chinese Translation.

From the emergence of the concept of the Internet of Things to today, China has always been very concerned about the development of the Internet of Things technology. The research and application of Internet of Things technology has become an important measure for the country to promote supply-side reform. The Internet of Things technology is the core innovation content of the country and an important measure to realize the smart city. This paper analyzes the cutting-edge research results in the field of Internet of Things abroad. This paper focuses on the application of IoT technology in English terminology research. This paper specifically analyzes the feature extraction methods of Internet of Things terms and studies the application content of Internet of Things technology in English-Chinese translation. We specifically conduct translation research on the content in “The Internet of Things: Enabling Technologies, Platforms, and Use Cases,” focusing on the translation of scientific and technological terms. As a kind of technical English terminology, IoT English terminology has the characteristics of strong professionalism and rigorous semantics. This article specifically translates the original part of the practice report and classifies the terms that appear in it, including IoT English terms for which canonical translations have been made and IoT English terms that do not have standardized translations. The research will further carry out the investigation and analysis work. The research is to identify the terms that have standardized translation, especially the detailed research of the terms of the industry field. This paper selects standard translations and summarizes the translation methods of terms from three aspects: abbreviations, compound words, and semi-technical words. The research results provide guidance for terminology translation. This article translates terms that do not yet have a canonical translation. Based on terminology characteristics and existing terminology translation methods, this paper proposes a variety of translation methods such as literal translation, decomposing and combining, and non-translation. The study described in detail with examples. The research results of this paper are expected to provide some reference for translators engaged in the translation of scientific and technological documents.

A coarse-to-fine full attention guided capsule network for medical image segmentation

As an important material in clinical diagnoses, medical images are widely used by the doctors to discover the diseases and make therapeutic schedules. Accurately locating the lesions and correctly delineating their severities based on the medical images can significantly help to improve the detection rate and diagnosis accuracy of the diseases. In this paper, we design an effective attention guided capsule network, named HR-CapsSegNet, for segmenting medical targets from medical images. First, by forming a capsule-based high-resolution network architecture assisted by cross-branch multiscale feature augmentation, the HR-CapsSegNet performs advantageously in providing multiscale feature representations with high-level strong semantics. Second, by designing a capsule-based full attention mechanism, the feature encoding quality at each scale can be significantly promoted by sticking up the informative feature semantics from a cross-channel global perspective. In addition, by employing a hierarchical coarse-to-fine segmentation strategy, the feature distractions causing false recognitions can be progressively removed to provide an accurate segmentation map. Intensive quantitative assessments, visual examinations, and comparative analyses on four challenging datasets prove the promising applicability and competitive superiority of the HR-CapsSegNet for medical image segmentation applications.

From enhanced coinduction towards enhanced induction

There exist a rich and well-developed theory of enhancements of the coinduction proof method, widely used on behavioural relations such as bisimilarity. We study how to develop an analogous theory for inductive behaviour relations, i.e., relations defined from inductive observables. Similarly to the coinductive setting, our theory makes use of (semi)-progressions of the form R->F(R), where R is a relation on processes and F is a function on relations, meaning that there is an appropriate match on the transitions that the processes in R can perform in which the process derivatives are in F(R). For a given preorder, an enhancement corresponds to a sound function, i.e., one for which R->F(R) implies that R is contained in the preorder; and similarly for equivalences. We introduce weights on the observables of an inductive relation, and a weight-preserving condition on functions that guarantees soundness. We show that the class of functions contains non-trivial functions and enjoys closure properties with respect to desirable function constructors, so to be able to derive sophisticated sound functions (and hence sophisticated proof techniques) from simpler ones. We consider both strong semantics (in which all actions are treated equally) and weak semantics (in which one abstracts from internal transitions). We test our enhancements on a few non-trivial examples.

Knowledge graph analytics platform with LINCS and IDG for Parkinson's disease target illumination

BackgroundLINCS, "Library of Integrated Network-based Cellular Signatures", and IDG, "Illuminating the Druggable Genome", are both NIH projects and consortia that have generated rich datasets for the study of the molecular basis of human health and disease. LINCS L1000 expression signatures provide unbiased systems/omics experimental evidence. IDG provides compiled and curated knowledge for illumination and prioritization of novel drug target hypotheses. Together, these resources can support a powerful new approach to identifying novel drug targets for complex diseases, such as Parkinson's disease (PD), which continues to inflict severe harm on human health, and resist traditional research approaches.ResultsIntegrating LINCS and IDG, we built the Knowledge Graph Analytics Platform (KGAP) to support an important use case: identification and prioritization of drug target hypotheses for associated diseases. The KGAP approach includes strong semantics interpretable by domain scientists and a robust, high performance implementation of a graph database and related analytical methods. Illustrating the value of our approach, we investigated results from queries relevant to PD. Approved PD drug indications from IDG’s resource DrugCentral were used as starting points for evidence paths exploring chemogenomic space via LINCS expression signatures for associated genes, evaluated as target hypotheses by integration with IDG. The KG-analytic scoring function was validated against a gold standard dataset of genes associated with PD as elucidated, published mechanism-of-action drug targets, also from DrugCentral. IDG's resource TIN-X was used to rank and filter KGAP results for novel PD targets, and one, SYNGR3 (Synaptogyrin-3), was manually investigated further as a case study and plausible new drug target for PD.ConclusionsThe synergy of LINCS and IDG, via KG methods, empowers graph analytics methods for the investigation of the molecular basis of complex diseases, and specifically for identification and prioritization of novel drug targets. The KGAP approach enables downstream applications via integration with resources similarly aligned with modern KG methodology. The generality of the approach indicates that KGAP is applicable to many disease areas, in addition to PD, the focus of this paper.

ForkNet: Strong Semantic Feature Representation and Subregion Supervision for Accurate Remote Sensing Change Detection

In this article, we propose an effective siamese feature pyramid network (FPN), ForkNet, for remote sensing change detection (RSCD). We find that the siamese network structure, which is widely used for RSCD, contains only one downsampling network in the feature extraction stage, e.g., VGG16 and ResNet-18, to extract the deep features of a single image, such that the features have a large semantic gap between high-level feature maps and low-level feature maps. The low-level feature maps with weak semantics may become a bottleneck of network performance. Thus, we apply an FPN to the feature extraction stage to generate feature representations with strong semantics at each level. Further, we design a cross-resolution attention module (CRAM) to aggregate contextual information across resolutions and naturally serve as a bridge for exchanging information across different resolution feature maps. The siamese FPN equipped with the CRAM is called ForkNet. To better train ForkNet, we extend the Tversky loss to a novel loss, pyramid Tversky loss, which is capable of supervising subregions at different scales to obtain more fine-grained detection results. Using pyramid Tversky loss together with Focal loss, our ForkNet achieves state-of-the-art detection performance on two challenging datasets.

Assume-guarantee contracts for continuous-time systems

Many techniques for verifying properties of continuous-time systems are limited to systems of moderate size. In this paper, we propose an approach based on assume-guarantee contracts and compositional reasoning for verifying properties of a broad class of continuous-time systems consisting of interconnected components. The notion of assume-guarantee contracts makes it possible to divide responsibilities among the system components: a contract specifies the property that a component must fulfill under some assumptions on the behavior of its environment (i.e. of the other components). We define weak and strong semantics of assume-guarantee contracts. We then establish a certain number of results for compositional reasoning, which allow us to show that a global assume-guarantee contract of the whole system is satisfied when all components satisfy their own contracts. We show that the weak satisfaction of the contract is sufficient to deal with interconnections described by a directed acyclic graph, while strong satisfaction is needed to reason about general interconnections containing cycles. Specific results for systems described by differential inclusions and invariance assume-guarantee contracts are then developed. Finally, we show how the proposed assume-guarantee framework can recast different versions of the small-gain theorem as a particular case. Throughout the paper, the main results are illustrated using simple examples.

SSWS: An edge detection algorithm with strong semantics and high detectability for spacecraft

Space tasks such as docking, capture and maintenance of spacecraft in orbit are increasingly dependent on visual system. High-order features extraction, target recognition and spacecraft pose measurement are essential image processing technologies to accomplish above tasks, for which image edge detection is one of the most critical and basic techniques. Traditional edge detection algorithms such as Canny have good extraction effect, but cannot extract and obtain semantic information. Besides, most of the edge detection algorithms based on deep-learning rely on a large number of annotation labels and require huge workload. Therefore, they cannot be applied to non-cooperative space targets. In this paper, we propose an edge detection algorithm with strong semantics and weak supervision (SSWS) to combine edge detection with image semantic segmentation. By learning geometric boundaries of ellipse and quadrilateral for docking ring and solar panel of spacecraft, and using the training method of weak supervision, our algorithm achieves effective boundary detection of important parts of spacecraft from images regardless of the background interference and light intensity. Real image experiments show that SSWS can effectively extract the boundary of important parts of spacecraft and filter out the background information and noise. In the performance verification experiment of BSDS500 dataset, with weakly supervised annotation tags, SSWS obtains the best OIS (0.831), in comparison with traditional edge detection algorithms and excellent recent deep-learning edge detection algorithms in recent years; in the contrast experiment of image segmentation in PASCAL VOC 2012 test set, SSWS gets the best mIoU value (66.9%).

'Believe' is Strong but Subjective: Experimental Evidence from Hedging

This paper contributes to the debate concerning whether the attitude verb believe has a weak or a strong semantics. According to Hawthorne et al. (2016), believe is weak and akin to the probability operator likely, typically receiving an “agent finds it more likely than not” interpretation. Alternatively, Koev (2019) proposes that believe conveys high certainty but qualifies this certainty as subjective or lacking evidence, in contrast with modals that convey high objective certainty, like sure (see also Lyons 1977; Kratzer 1981; Nuyts 2001; Papafragou 2006; Portner 2009). We focus on the use of believe as a hedge (e.g. I believe the Giants will win, but I’m not sure they will) as allegedly the most convincing argument for the weak view, and argue that, in fact, it favors the strong-but-subjective view. We show experimentally that the availability of the hedging use of believe is affected by certain grammatical and discourse factors. Experiment 1 reveals that participants rate hedging sentences with combinations of third person/past tense/embedded features as less natural than canonical first person/present tense/main clause forms. In turn, Experiment 2 reveals that hedging sentences with at-issue prejacents are judged as more natural than sentences in which the belief component is at-issue. The observed variability posits a challenge to the weak view, which establishes a purely logical contrast in modal strength between likelihood vs. certainty. However, it is in line with the strong-but-subjective view, which establishes a contrast in modal content between certainty without evidence vs. certainty with evidence and predicts a more restricted distribution of the hedging reading.

EnvelopeNet: A robust convolutional neural network with optimal kernels for intelligent fault diagnosis of rolling bearings

Deep data-driven methods for fault diagnosis, as an engineering-oriented approach, rely heavily on target data. For engineering applications, the working conditions of rotating machinery fluctuate from time to time and a collection for any working conditions is impossible. To tackle this problem, a robust network with optimal kernels named EnvelopeNet is proposed for extracting solid information and eliminating the influence of fluctuations. In the EnvelopeNet, a feature evaluation building block named envelope module is constructed based on optimal band selection theory to optimize the kernels. Compared with the kurtogram, the learned optimal kernels and features show strong semantics which helps the network become robust. The EnvelopeNet is validated under the approximate speed and mixed speed scenarios. The results show that the EnvelopeNet could provide admirable generalization ability for fluctuating working conditions and an average improvement of about 4% and 3% over existing approaches under two scenarios respectively.

A Framework for Ontology Based Semantic Search System in Ayurvedic Medicine

India is known for its traditional medicinal system such as Ayurveda, Yoga, Unani, Siddha and Homeopathy. Ayurveda plays a significant role in curing the diseases without any side effects. Medicinal plants or herbs are considered as a major resource in meeting the need of people health care. Information about this medicinal knowledge must be preserved and digitized. There have been a massive number of publications and large number of articles on ayurvedic research in the form of unstructured textual data. Text mining approach is used to provide the solution to handle such voluminous of unstructured data. With the exponential growth of text based data, navigating the relevant information needed is the challenging task. Semantic understanding of document content forms the vital requirement for ensuring the quality of content retrieval. However, the current approaches are finding variation in textual classification in bringing the classification accuracy which may fail to understand the data during classification. Hence, an efficient model is required to search, classify and retrieve the most relevant data. The main objective of this research is to develop an effective and efficient framework and algorithm to search and retrieve the most relevant facts by including the application of ontology-based text mining approach. The current status of research is analyzed and reviewed in the area of semantic web retrieval, ontology-based approaches and various classification technique for building the framework. Text mining with the special emphasis on understanding the semantic meaning of content is achieved by using domain ontology called medicinal plant ontology construction. The challenges in finding the semantically related content for the given query are achieved through semantic web and ontology which enriched the data on web for structured representation thereby providing the strong semantics in knowledge representation. The methodology of information extraction is implemented by using medicinal plant ontology with semantic knowledge representation, an algorithm called OCEC (Ontology based Concept Extraction and Classification) was developed where each term is described semantically by mapping the terms and its related terms in the medicinal plant ontology. The web language called Web Ontology Language (OWL) is used for knowledge representation and is considered as richer semantic description language for describing unstructured and semi-structured content on the web thereby extracting the exact and relevant data and to offer a strong semantic search. To evaluate the performance of the proposed method, less relevant and most relevant documents were collected from online sources and digital libraries. Comparative study has been performed with various classification techniques. The experimental results show that the proposed method out performed. To further prove the efficiency of the model, experiments were conducted by giving different queries and the results are compared with other existing methods. The results show that the content retrieved by the proposed model improves precision and recall results. Keyword : Traditional Medicine, Ayurveda, Ontology, Semantic Web, Web Ontology Language