Human-inspired semantic similarity between sentences

Wikipedia bi-linear link (WBLM) model: A new approach for measuring semantic similarity and relatedness between linguistic concepts using Wikipedia link structure

In natural language processing (NLP) tasks, computing semantic textual similarity (STS) is crucial for capturing nuanced semantic differences in text. Traditional word vector methods, such as Word2Vec and GloVe, as well as deep learning models like BERT, face limitations in handling context dependency and polysemy and present challenges in computational resources and real-time processing. To address these issues, this paper introduces two novel methods. First, a sentence embedding generation method based on Kullback–Leibler Divergence (KLD) optimization is proposed, which enhances semantic differentiation between sentence vectors, thereby improving the accuracy of textual similarity computation. Second, this study proposes a framework incorporating RoBERTa knowledge distillation, which integrates the deep semantic insights of the RoBERTa model with prior methodologies to enhance sentence embeddings while preserving computational efficiency. Additionally, the study extends its contributions to sentiment analysis tasks by leveraging the enhanced embeddings for classification. The sentiment analysis experiments, conducted using a Stochastic Gradient Descent (SGD) classifier on the ACL IMDB dataset, demonstrate the effectiveness of the proposed methods, achieving high precision, recall, and F1 score metrics. To further augment model accuracy and efficacy, a feature selection approach is introduced, specifically through the Dynamic Principal Component Selection (DPCS) algorithm. The DPCS method autonomously identifies and prioritizes critical features, thus enriching the expressive capacity of sentence vectors and significantly advancing the accuracy of similarity computations. Experimental results demonstrate that our method outperforms existing methods in semantic similarity computation on the SemEval-2016 dataset. When evaluated using cosine similarity of average vectors, our model achieved a Pearson correlation coefficient (τ) of 0.470, a Spearman correlation coefficient (ρ) of 0.481, and a mean absolute error (MAE) of 2.100. Compared to traditional methods such as Word2Vec, GloVe, and FastText, our method significantly enhances similarity computation accuracy. Using TF-IDF-weighted cosine similarity evaluation, our model achieved a τ of 0.528, ρ of 0.518, and an MAE of 1.343. Additionally, in the cosine similarity assessment leveraging the Dynamic Principal Component Smoothing (DPCS) algorithm, our model achieved a τ of 0.530, ρ of 0.518, and an MAE of 1.320, further demonstrating the method’s effectiveness and precision in handling semantic similarity. These results indicate that our proposed method has high relevance and low error in semantic textual similarity tasks, thereby better capturing subtle semantic differences between texts.

A semantic similarity computation method for virtual resources in cloud manufacturing environment based on information content

The sharing and interoperation of geographical information has always been a hot topic in the field of geographical information science, and the semantic similarity of the concept of geographic information is also an important factor affecting geographic information sharing and interoperability. The semantic similarity of concept calculation is a natural language processing, an important component of the study, but also applications of artificial intelligence problems to be solved. Ontology, which is an explicit and formal specification of sharing information, becoming an effective method for solve the problem of semantic similarity. Considering the special nature of the concept of geographic information, this paper takes an integrated multi-point semantic similarity calculation method in this paper. We made an investigation on the distance-based semantic similarity calculation model, the attribute-based semantic similarity calculation model, the relationship between concepts-based semantic similarity calculation models. And we analyzed all aspects of the concept of geographic information. Then, considering the impact of the distance, the attributes, the relationship between concepts for the semantic similarity of the concept of geographic information, this paper established an integrated and multi-angle model for the semantic similarity calculation of geographic information concept. And we added some adjustable parameters to adapt the flexible application. Make sure the calculation results more accurate and objective. By precise experimental analysis, this integrated model can improve the calculation accuracy.

In the field of biology, researchers need to compare genes or gene products using semantic similarity measures (SSM). Continuous data growth and diversity in data characteristics comprise what is called big data; current biological SSMs cannot handle big data. Therefore, these measures need the ability to control the size of big data. We used parallel and distributed processing by splitting data into multiple partitions and applied SSM measures to each partition; this approach helped manage big data scalability and computational problems. Our solution involves three steps: split gene ontology (GO), data clustering, and semantic similarity calculation. To test this method, split GO and data clustering algorithms were defined and assessed for performance in the first two steps. Three of the best SSMs in biology [Resnik, Shortest Semantic Differentiation Distance (SSDD), and SORA] are enhanced by introducing threaded parallel processing, which is used in the third step. Our results demonstrate that introducing threads in SSMs reduced the time of calculating semantic similarity between gene pairs and improved performance of the three SSMs. Average time was reduced by 24.51% for Resnik, 22.93%, for SSDD, and 33.68% for SORA. Total time was reduced by 8.88% for Resnik, 23.14% for SSDD, and 39.27% for SORA. Using these threaded measures in the distributed system, combined with using split GO and data clustering algorithms to split input data based on their similarity, reduced the average time more than did the approach of equally dividing input data. Time reduction increased with increasing number of splits. Time reduction percentage was 24.1%, 39.2%, and 66.6% for Threaded SSDD; 33.0%, 78.2%, and 93.1% for Threaded SORA in the case of 2, 3, and 4 slaves, respectively; and 92.04% for Threaded Resnik in the case of four slaves.

Sentence similarity computation is a key part of the Chinese FAQ system and the computation accuracy determine the quality of the system. Combining with question classification index, a new Chinese sentence similarity computing method based on Content Forward Network is proposed. The method combined concept semantic matching and the overall semantic question sentence similarity computation with question automatic classification to deal with the sentence similarity computing. Comparing with the traditional keyword-based and simple semantic question similarity computation method, Chinese frame semantic information makes the Chinese sentence similarity calculation accuracy improved obviously. The experimental results show that the method can effectively solve the intelligent matching problem in Chinese question answering system. Keywords—Sentence similarity; Chinese Frame Network; Frame element; Question classification; Question answering system

Document summarization provides an instrument for faster understanding the collection of text documents and has a number of real life applications. Semantic similarity and clustering can be utilized efficiently for generating effective summary of large text collections. Summarizing large volume of text is a challenging and time consuming problem particularly while considering the semantic similarity computation in summarization process. Summarization of text collection involves intensive text processing and computations to generate the summary. MapReduce is proven state of art technology for handling Big Data. In this paper, a novel framework based on MapReduce technology is proposed for summarizing large text collection. The proposed technique is designed using semantic similarity based clustering and topic modeling using Latent Dirichlet Allocation (LDA) for summarizing the large text collection over MapReduce framework. The summarization task is performed in four stages and provides a modular implementation of multiple documents summarization. The presented technique is evaluated in terms of scalability and various text summarization parameters namely, compression ratio, retention ratio, ROUGE and Pyramid score are also measured. The advantages of MapReduce framework are clearly visible from the experiments and it is also demonstrated that MapReduce provides a faster implementation of summarizing large text collections and is a powerful tool in Big Text Data analysis.

Semantic textual similarity is the basis of countless applications and plays an important role in diverse areas, such as information retrieval, plagiarism detection, information extraction and machine translation. This article proposes an innovative word embedding-based system devoted to calculate the semantic similarity in Arabic sentences. The main idea is to exploit vectors as word representations in a multidi-mensional space in order to capture the semantic and syntactic properties of words. IDF weighting and Part-of-Speech tagging are applied on the examined sentences to support the identification of words that are highly descriptive in each sentence. The performance of our proposed system is confirmed through the Pearson correlation between our assigned semantic similarity scores and human judgments.

Geographic entity matching is an important means for multi-source spatial data fusion and information association and sharing. Corresponding matching methods have been designed by existing studies for different types of entity data characteristics, such as line and area. However, these approaches are often limited in the generalization ability for matching heterogeneous data from multiple sources and the accuracy for complex pattern matching. To resolve these problems, robust multi-source geographic entities matching by maximizing geometric and semantic similarity is proposed. First, the entire entity is segmented based on shape features, and the partitioned feature segments are extracted as matching primitives; Second, feature segments are grouped into patterns, encompassing three major categories and fourteen subcategories; Following this, pattern matching is performed based on spatial similarity metric such as maximum projection distance, etc.; Finally, the spatial matches are detected and refined through semantic similarity calculation. The proposed method is tested using two datasets from regions in southeast and northwest China. The experimental results demonstrate that our method can be effectively applied to both area and line entity matching with strong generalization and application capability and significantly improved matching accuracy. Specifically, nine feature segment matching patterns for matching area entities and six for line entities are utilized, and the precision and recall are nearly 90%.

With the increasing complexity of scientific and technological project management, traditional project management systems are faced with the challenges of massive data processing and semantic understanding. The study first discusses the computer-implemented technology of semantic similarity evaluation, analyzes the application of word embedding and deep learning algorithms, and emphasizes the importance of these technologies in natural language processing (NLP). A novel hybrid semantic similarity evaluation framework has been developed and implemented, incorporating an optimized algorithm-based semantic feature extraction module alongside multi-source data fusion and processing techniques modeled on semantic relationships. The framework’s primary objective is to enable precise analysis of multilingual project data by deeply exploring contextual information. Experimental validation highlights the framework’s performance and effectiveness, with particular emphasis on its accuracy in semantic similarity calculations, data processing efficiency, and responsiveness in large-scale environments. During testing, the framework achieved a semantic similarity calculation accuracy of 67.37%, based on a test set comprising 73 samples. An analysis of 74,723 project records revealed that the system’s average response time for multi-source data fusion was 2.3 seconds. Under complex semantic analysis conditions, the framework processed data at a rate of 348.15 items per second. Multiple test results further demonstrated that the error margin in semantic evaluation remained within 2.51 percentage points. These findings confirm that the system optimization successfully meets the anticipated goals in terms of both accuracy and efficiency.

BackgroundOntologies are widely used as metadata in biological and biomedical datasets. Measures of semantic similarity utilize ontologies to determine how similar two entities annotated with classes from ontologies are, and semantic similarity is increasingly applied in applications ranging from diagnosis of disease to investigation in gene networks and functions of gene products.ResultsHere, we analyze a large number of semantic similarity measures and the sensitivity of similarity values to the number of annotations of entities, difference in annotation size and to the depth or specificity of annotation classes. We find that most similarity measures are sensitive to the number of annotations of entities, difference in annotation size as well as to the depth of annotation classes; well-studied and richly annotated entities will usually show higher similarity than entities with only few annotations even in the absence of any biological relation.ConclusionsOur findings may have significant impact on the interpretation of results that rely on measures of semantic similarity, and we demonstrate how the sensitivity to annotation size can lead to a bias when using semantic similarity to predict protein-protein interactions.

Text similarity is one of the important methods of text data analysis, which is often used in text clustering and classification. Social media is a new and popular online social application that contains a lot of valuable information. Short text is common in social media, and short text similarity is often used for social media data mining. The similarity calculation of short text is influenced by the small feature of text words and the accuracy is low. so it is a common improvement method to calculate the similarity of short texts with word semantic similarity. This paper put forward a short text semantic similarity calculation method that combine knowledge-based method and corpus-based method. This method is based on the improved word semantic similarity calculation method and general short text semantic similarity calculation method. The word similarity calculation method combines two word semantic similarity by some strategies. It takes the advantages of two methods to overcome the disadvantages of single one, finds out more semantic association among words in texts, and improves accuracy of word similarity calculation. This paper uses a large number of corpus to compare and analyze several word and text semantic similarity algorithms, the improved method has a closer result to human ratings than other methods in both word and text similarity.

Semantic similarity is a useful approach for comparing patient phenotypes, and holds the potential of an effective method for exploiting text-derived phenotypes for differential diagnosis, text and document classification, and outcome prediction. While approaches for context disambiguation are commonly used in text mining applications, forming a standard component of information extraction pipelines, their effects on semantic similarity calculations have not been widely explored. In this work, we evaluate how inclusion and disclusion of negated and uncertain mentions of concepts from text-derived phenotypes affects similarity of patients, and the use of those profiles to predict diagnosis. We report on the effectiveness of these approaches and report a very small, yet significant, improvement in performance when classifying primary diagnosis over MIMIC-III patient visits.

Domain words clustering have important theoretical and practical significance in text categorization, the ontology research, machine learning and many other research areas. The domain words clustering method in this article is a method based on word2vec and semantic similarity computation. First of all, we get the candidate word set with word2vec tools to preliminary clustering of words. Then we tectonic domain category semantic core word set and screening candidate word set by means of semantic similarity computation. Finally we get new word set belongs to the target domain and get the word set in the field of clustering. Experiments show that this method has higher recall ratio and accuracy.

In recent years, the exploitation of sentence embedding techniques in natural language processing field has encouraged the proposal of new methods for extracting keyphrases from documents based on these techniques. Most of these approaches select keyphrases from a set of candidate phrases based on their semantic proximity to the document. In general, most documents contain complementary paragraphs that are unrelated to the topics covered. This factor reduces the credibility of the semantic proximity of candidate keyphrases to the document. Exploitation of document paragraphs weights during the semantic similarity calculation will inevitably improve the performance of keyphrase extraction from document. In this paper, we propose a new method to extract keyphrases based on document paragraphs weights. Our method is based on sentence embedding techniques and semantic proximity of candidate key phrases from document paragraphs. We evaluated the proposed method on three datasets, Inspec, Semeval2010 and KPTimes, where our results showed that that using document paragraph weight improved the performance of keyphrases extraction.