Optimization of science and technology project management system based on hybrid semantic similarity evaluation framework

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 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.

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%.

Natural language processing (NLP) is a critical research direction in artificial intelligence, where the calculation of vocabulary semantic similarity is the foundation and core work. However, existing calculation methods are faced with problems, e.g., the inability to extract important semantic information. Failure to address this issue can compromise the accuracy of semantic similarity measures in NLP applications. To this end, in this paper, a vocabulary semantic similarity calculation model based on word vectors and convolutional neural networks (CNNs) was proposed. The word vector model was improved using long short-term memory (LSTM) networks, and important semantics were extracted using convolutional layers and ensured semantic order through bidirectional Gated Recurrent Unit. The structure of the Siamese neural network was used to ensure consistency in text encoding. The experimental findings have shown that the proposed model has the highest F1 value in different datasets. In the original Chinese natural language inference (OCNLI) dataset, the Pearson correlation coefficient of the model was 0.021 and 0.018 higher than that of the LSTM network and CNNs, respectively. The accuracy of the similarity calculation in the two datasets was 92.4% and 96.5%, respectively. According to these results, the semantic similarity prediction value of the proposed model can be closer to the true value, and the prediction performance of the model is more excellent.

Semantic similarity is an essential concept that widen across various fields such as artificial intelligence, natural language processing, information retrieval, relation extraction, document clustering and automatic data extraction. The study proposed in this paper focus on semantic similarity and counter measure on question answering task. Despite the usefulness of semantic similarity in these applications, finding the exact meaning between the words from question and answer pair has become a major challenge in question answering system. To recognize various relationships those exist between question and answer pair, measuring semantically is essential. We present study using a match maker algorithm for similarity measurement between question-answer pair. The study compares similarity measurement using match maker algorithm as against semantic similarity measure on Miller-Charles bench marked data set. This paper present study about different types of algorithm used to measure the semantic similarity between the words and those result compare with match maker algorithm. The experiments on real datasets shows matchmaker algorithm works better than web-based semantic similarity measure.

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

Big data analytics (BDA) is an advanced analytic technique used with very large and diverse sets of data from different sources. Natural language processing (NLP) is a technology that interfaces with different fields such as computer science, linguistics, and human-computer interactions. Over the past few years, there is a growing number of firms, which are using different BDA and NLP applications in their businesses. Only a few of the research have investigated different dimensions of NLP and BDA and their impacts on the overall organizational performance. There is a growing interest among researchers and practitioners in understanding the consequences for firms that adopt BDA and NLP applications. In this context, the aim of this article is to determine the factors for the usage of BDA and NLP applications in business. With the help of dynamic capability view theory and existing literature, a theoretical model was developed conceptually. Later, the model was validated using structural equation modeling approach considering 1287 samples from 23 firms, primarily based in Asia and Europe, which use NLP and BDA applications. The article finds that NLP and BDA applications help the firms to improve their operational efficiency, which in turn improves the overall firm performance.

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

The rise of homogenization and polarization in the news may inhibit individuals’ understanding of an issue and the functioning of a democratic society. This study applies a network approach to understanding patterns of semantic similarity and divergence across news coverage. Specifically, we focus on how (a) inter-organizational networks based on media ideology, (b) inter-organizational networks based on news truthfulness, and (c) public engagement that news articles received on social media may affect semantic similarity in the news. We use large-scale user logs data on social media platforms (i.e., Facebook and Twitter) and news text data from more than 100 news organizations over 10 months to examine the three potential processes. Our results show that the similarity between news organizations in terms of media ideology and news truthfulness is positively associated with semantic similarity, whereas the public engagement that news articles received on social media is negatively associated with semantic similarity. Our study contributes to theory development in mass communication by shifting to a network paradigm that connects news organizations, news content, and news audiences. We demonstrate how scholars across communication disciplines may collaborate to integrate distinct theories, connect multiple levels, and link otherwise separate dimensions. Methodologically, we demonstrate how synchronizing network science with natural language processing and combining social media log data with text data can help to answer research questions that communication scholars are interested in. The findings’ implications for news polarization are discussed.

With the rapid development of the power grid, relevant personnel have increasingly high requirements for the safety and stability of the power grid. Digital video surveillance technology is widely used in the safety monitoring and fault detection of power grids. However, traditional video surveillance systems have problems such as limited monitoring range, low processing efficiency, and excessive manual intervention. The article studied a digital video monitoring platform for smart grids based on the DL (Deep Learning) algorithm. The DL algorithm is used for face detection to identify face information in the video stream; through data processing and feature extraction, facial features are analyzed and compared; the DL algorithm is used for face recognition to achieve accurate identification and monitoring of individuals in the power grid environment, thereby improving safety and management efficiency. This platform can effectively improve the security (Max: 95.2%) and efficiency (Max: 89%) of power grid monitoring. This platform has broad application prospects and can provide strong technical support and decision-making basis for the power grid industry.

Semantic Textual Similarity (STS), which measures the equivalence of meanings between two textual segments, is an important and useful task in Natural Language Processing. In this article, we have analyzed the datasets provided by the Semantic Evaluation (SemEval) 2012–2014 campaigns for this task in order to find out appropriate linguistic features for each dataset, taking into account the influence that linguistic features at different levels (e.g. syntactic constituents and lexical semantics) might have on the sentence similarity. Results indicate that a linguistic feature may have a different effect on different corpus due to the great difference in sentence structure and vocabulary between datasets. Thus, we conclude that the selection of linguistic features according to the genre of the text might be a good strategy for obtaining better results in the STS task. This analysis could be a useful reference for measuring system building and linguistic feature tuning.

The PAP900 dataset centers on the semantic relationship between affective words in Portuguese. It contains 900 word pairs, each annotated by at least 30 human raters for both semantic similarity and semantic relatedness. In addition to the semantic ratings, the dataset includes the word categorization used to build the word pairs and detailed sociodemographic information about annotators, enabling the analysis of the influence of personal factors on the perception of semantic relationships. Furthermore, this article describes in detail the dataset construction process, from word selection to agreement metrics.Data was collected from Portuguese university psychology students, who completed two rounds of questionnaires. In the first round annotators were asked to rate word pairs on either semantic similarity or relatedness. The second round switched the relation type for most annotators, with a small percentage being asked to repeat the same relation. The instructions given emphasized the differences between semantic relatedness and semantic similarity, and provided examples of expected ratings of both.There are few semantic relations datasets in Portuguese, and none focusing on affective words. PAP900 is distributed in distinct formats to be easy to use for both researchers just looking for the final averaged values and for researchers looking to take advantage of the individual ratings, the word categorization and the annotator data. This dataset is a valuable resource for researchers in computational linguistics, natural language processing, psychology, and cognitive science.

This research focuses on optimizing the speed of Big Data processing using Artificial Intelligence (AI) in healthcare applications. The study integrates Random Forest (RF) and Deep Learning (DL) algorithms with cloud-based computing systems to improve data processing efficiency. The dataset includes both structured data, such as Electronic Health Records (EHR), and unstructured data, like medical images. The results show that RF performs better with structured data, achieving a lower Mean Squared Error (MSE) and higher R-squared (R²) than traditional methods. Meanwhile, DL achieves superior accuracy and Area Under the Curve (AUC) in processing unstructured data. By utilizing the distributed computing power of Spark on a cloud platform, the processing speed was significantly enhanced, as demonstrated by a statistically significant reduction in processing time (p < 0.05) observed through a t-test analysis comparing Spark-based computing with traditional methods. Despite these improvements, challenges such as data privacy and infrastructure costs remain. Despite these improvements, challenges such as data privacy and infrastructure costs remain. This research provides a robust framework for real-time healthcare data analysis, highlighting its potential to improve decision-making processes and patient outcomes in medical services.

Human-inspired semantic similarity between sentences

Humans’ most powerful tool is their mental wellness. Individuals’ well-being can be impacted by poor mental health. This paper focuses on a smart technical solution to the problem of mental health issues detection related to the stress, sadness, depression, anxiety, etc. which if not handled efficiently may further lead to a severe problem. The paper deals with the designing of an automated smart system using social media posts that will help mental health experts to successfully identify and understand about the mental health condition of social media users. That can be done based on text analysis of rich social media resources such as Reddit, Twitter posts. The implementation of the system is done using Natural Language Processing (NLP) methods, machine learning and deep learning algorithms. The models are trained using a prepared dataset of social media postings. With this automated system the mental health experts can able to detect the stress or some other emotions of social media uses in a very earlier as well as faster way. The proposed system can predict five emotional categories: ‘Happy’, ‘Angry’, ‘Surprise’, ‘Sad’, ‘Fear’ based on machine learning (Logistic Regression, Random Forest, SVM), deep learning Long Short-Term Memory (LSTM) and BERT transfer learning algorithms. All the applied algorithms are evaluated using confusion matrix, the highest accuracy and f1 score achieved is more than 85%, which is better than the existing human emotion detection systems.KeywordsNatural language processing (NLP)Text analysisMachine learningDeep learningTransfer learningText2emotionSocial media postsLSTMBERT