Structured Knowledge Research Articles

K-PathVQA: Knowledge-Aware Multimodal Representation for Pathology Visual Question Answering.

Pathology imaging is routinely used to detect the underlying effects and causes of diseases or injuries. Pathology visual question answering (PathVQA) aims to enable computers to answer questions about clinical visual findings from pathology images. Prior work on PathVQA has focused on directly analyzing the image content using conventional pretrained encoders without utilizing relevant external information when the image content is inadequate. In this paper, we present a knowledge-driven PathVQA (K-PathVQA), which uses a medical knowledge graph (KG) from a complementary external structured knowledge base to infer answers for the PathVQA task. K-PathVQA improves the question representation with external medical knowledge and then aggregates vision, language, and knowledge embeddings to learn a joint knowledge-image-question representation. Our experiments using a publicly available PathVQA dataset showed that our K-PathVQA outperformed the best baseline method with an increase of 4.15% in accuracy for the overall task, an increase of 4.40% in open-ended question type and an absolute increase of 1.03% in closed-ended question types. Ablation testing shows the impact of each of the contributions. Generalizability of the method is demonstrated with a separate medical VQA dataset.

Impact of a nurse-led intervention in understanding menstrual health management among adolescent girls attending specific schools in Navi Mumbai, India

Abstract Background: Premenstrual symptoms are prevalent among menstruating individuals, affecting around 75% of those who report some level of discomfort during their cycles. However, approximately 10% of women endure premenstrual syndrome (PMS), marked by symptoms that significantly disrupt daily functioning. PMS remains poorly comprehended and difficult to diagnose. The aim of this study was to assess the efficacy of an intervention and analyze the disparity in knowledge levels between the two groups. Materials and Methods: A genuine experimental design was employed using a pretest and post-test control group approach. A simple random sampling lottery method was used to select 40 adolescent girls for the study and control groups. These girls, aged 12–15 years, were chosen from specific schools in Navi Mumbai, adhering to predetermined inclusion criteria. A structured knowledge questionnaire was administered to both groups. At 1 week post-test, the study group received a nurse-led intervention through a planned teaching program. Data analysis included descriptive and inferential statistics. Results: In the control group, pretest knowledge scores averaged around 57.5%, suggesting a moderate level of knowledge. Conversely, the study group exhibited an average pretest knowledge score of 60%. Examining post-test knowledge scores, it is evident that in the control group, the majority (42.5%) demonstrated average knowledge, while in the study group, 67.5% exhibited excellent knowledge. Paired t test findings indicated a significant influence of the planned teaching program in the study group compared with the control group, with a P value of 0.001 at a significance level. Conclusion: Adolescent girls in the selected schools lacked knowledge about menstrual health management. Consequently, the nurse-led intervention on menstrual health management significantly improved knowledge among the control group of adolescent girls.

EFFECTIVENESS OF STRUCTURED TEACHING PROGRAM ON KNOWLEDGE REGARDING CYBERBULLYING AMONG 3RD YEAR B.SC. NURSING STUDENTS

Background of the study: This present study was done under the topic A study to assess the effectiveness of structured teaching program on knowledge regarding Cyberbullying among 3nd year B.Sc. Nursing students studying at SCPM College of nursing and paramedical sciences, Gonda. Methods: A Pre-experimental one group pretest posttest design was adopted for this study. 30 B.Sc. nursing students were selected as samples using Non probability purposive sampling technique. Socio demographic variables, Structured knowledge questionnaire were used as tools to assess the knowledge. Results: In pretest, 80.0% of the samples had inadequate knowledge score, 20.00% of them had moderate level of knowledge score and none of them had adequate knowledge score regarding Cyber bullying. Whereas in posttest, none of them had inadequate level knowledge score, 26.67% of them had moderate level of knowledge score and 73.33% of them are had adequate level of knowledge score regarding Cyber bullying. Statistically there was a significant difference between pretest and post-test knowledge score regarding cyber-bullying at the level of P< 0.001. Conclusion: This study concluded that structured teaching program was found to be an effective teaching strategy in improving knowledge of nursing students regarding cyber bullying.

A Study to assess the Effectivness of Planned Teaching Progamme on Dengue Fever in Terms of Knowledge

The study was experimental in nature with one group pre-test post-test design. The Planned teaching programme on dengue fever was developed by reviewing the extensive. Literature on dengue fever in terms of knowledge. The samples of 60 people were selected through probability simple random sampling technique. The instrument used for generating necessary data was a structured knowledge questionnaire. The obtained data was analysed by using both descriptive and inferential statistical in terms of Mean, Standard Deviation, ‘t’value. In the overall content area, mean gain scores were comparatively higher after exposure to planned teaching programme on dengue fever findings indicate that the planned teaching programme is a suitable and effective method of instruction. On the basis of findings, the following recommendation were made: The study can be replicated on a large sample, there by findings can be generalised for a large population.

Effectiveness of Structured Teaching Programme on Knowledge Regarding Millets and its Importance among Adolescents

This study entitled “Study to evaluate the effectiveness of structured teaching programme on knowledge regarding millets and its importance among adolescents in selected community area at Salempur” submitted in the year 2024. Aims: The objectives are to evaluate the pre-test knowledge regarding millets and its importance among adolescents; to evaluate the post-test knowledge regarding millets and its importance among adolescents; to evaluate the effectiveness of structured teaching programme on knowledge regarding millets and its importance among adolescents; to find out association between pre-test knowledge score with selected demographic variables among adolescents. Hypothesis: H1: There will be a significant difference between mean pre-test knowledge score than their mean post-test knowledge score regarding millets and its importance among adolescents. H2: There will be significant association between pre-test level of knowledge score and demographic variables. Design and Setting: Quasi - experimental approach with one group pre-test post-test design was adopted. Methods and Materials: 60 samples were selected by purposive sampling technique. A structured knowledge questionnaire was utilized to assess the knowledge of the samples. Results: The mean pre-test knowledge score of samples regarding was 11.95, whereas mean post-test knowledge score was 20.90 with a mean difference of 9.95 and SD pretest was 2.80 and post-test was 2.65. The calculated 't' value 38.34 was greater than tabulated 't'= 1.98 which was statistically proved at 0.05 level of significance. It revealed that the Structured Teaching Programme was effective in increasing knowledge among the participants. Conclusion: This indicates that the Structured Teaching Programme was effective to enhance the level of knowledge regarding millets and its importance among adolescents.

EFFECTIVENESS OF CAREGIVER MEDIATED PSYCHOSOCIAL INTERVENTIONS ON KNOWLEDGE, COPING OF CAREGIVERS OF CHILDREN WITH ATTENTION DEFICIT HYPERACTIVITY DISORDER AT SELECTED SETTINGS

Background: In India, prevalence of ADHD among adolescents is 7.1%. The prevalence of ADHD range for children aged 8-15 years was between 7.6% to 15%. There is a need to improve the knowledge and coping of the caregivers of children with ADHD, in order to enhance the level of knowledge on ADHD in caregivers and to promote the coping strategies of caregivers of children with ADHD. Aim and objective: To assess the effectiveness of the caregiver mediated psychosocial intervention on knowledge and coping of the caregivers of children with ADHD Materials and Methods: A quantitative approach with a quasi-experimental non-equivalent control group design was chosen for the study involving 30 caregivers who were selected using a non-probability purposive sampling technique. A structured knowledge questionnaire and Cravers brief cope inventory was used as the data collection tool. Results: It was revealed that in experimental group, the overall mean knowledge score of caregivers in pre-test was 6.80 knowledge and it was 11.20 in post-test with the mean difference of 4.40 and t value was found to be 7.06, which showed high statistical significance. In experimental group the overall mean coping score of caregivers, in pre-test was 60.33 and it was 82.20 in post-test with the mean difference of 21.87 and t value was found to be 14.97, which showed high statistical significance. Conclusion: The study concluded that the caregiver mediated psychosocial intervention was effective in enhancing knowledge and increasing the effective coping among caregivers of children with ADHD.

Knowledge regarding immunization among mothers of under-five children in Uttarakhand

Background: Immunization is one of the most impactful and cost-effective health investments globally that helps in reducing the burden of infectious diseases keeping children safe. Mothers are the major role players with regard to their children’s immunization. Methods: A descriptive cross-sectional study was conducted to assess knowledge regarding immunization among mothers of under-five children in the Doiwala block of Dehradun, Uttarakhand. A total of one hundred mothers of under-five children were conveniently selected through door-to-door survey. A structured knowledge questionnaire on under-five immunization was administered through the interview technique to assess the knowledge of the mothers. Results: Among 100 mothers of under-five children 13% had poor knowledge, 63% had average knowledge and 24% had good knowledge regarding under-five immunization. There was significant association between age, education status and socioeconomic status of mothers with knowledge score regarding under-five immunization. Conclusions: There is a strong need to increase awareness and knowledge about immunization among children; its benefits and importance. There is also a need to educate people especially mothers regarding harmful consequences of incomplete immunization of children.

A turning point few-shot learning for COVID-19 diagnosis

The current landscape of medical diagnostics grapples with a critical challenge posed by the limitations of existing meta-learning techniques in interpreting complex representations from limited labeled data, particularly evident in COVID-19 datasets. In response to this gap, our paper introduces a groundbreaking Turning Point (TP) methodology designed to enhance the interpretability of machine learning diagnostics, specifically addressing the shortcomings highlighted in conventional meta-learning approaches. Our Turning Point-based Few-Shot Learning (TPFSL) model goes beyond traditional FSL methods by embracing structured knowledge representation, departing from unstructured metric spaces. We found that our TPFSL model outperformed the state-of-the-art models by an average of 4.50% in 1-shot learning and 4.43% in 5-shot learning after conducting extensive benchmarking on the COVCT, SARSCOV2, and SIRM datasets. Across all datasets studied, TPFSL outperforms the ProtoNet benchmark in 1-shot classification by 12.966% and in 5-shot classification by 11.033%. The importance of TP density, the structure of the network, and placement in improving model performance have been shown by a thorough set of ablation investigations; with its revolutionary TPFSL model, which tackles the shortcomings of current meta-learning methods head-on, COVID-19 diagnostic procedures can be made more accurate and valid in clinical situations.

Knowledge graph question answering based on relational attention enhancement and feature cross-coding

Knowledge graph question answering aims to answer natural language questions using structured knowledge graph data. The key to achieving this is having a correct semantic understanding of the question phrases. Query graph generation is an important step for knowledge graph question answering systems to tackle complex questions. Unlike simple single-hop questions, complex questions often require reasoning between multiple triples to get the right answer due to multiple entities, relationships and constraints, making it difficult to generate correct query graphs. In previous studies, researchers have primarily focused on improving the extraction and representation of question features, neglecting the prior structural information implicated in the question itself. In this paper, we propose a question structure classifier to classify the question structure, and alleviate the noise interference in query graph through classification results. In the classifier, we strengthen the information about the question structure through the attention mechanism, while weakening the irrelevant information. Moreover, a query graph sorting module based on feature cross-coding is proposed to sort candidate paths in the query graph using fine-grained feature interaction between words. Extensive experiments are conducted on two public datasets (MetaQA and WebQuestionsSP) and the experimental results show that the proposed method outperforms other baselines.

Learning Hierarchical Prompt with Structured Linguistic Knowledge for Vision-Language Models

Prompt learning has become a prevalent strategy for adapting vision-language foundation models to downstream tasks. As large language models (LLMs) have emerged, recent studies have explored the use of category-related descriptions as input to enhance prompt effectiveness. Nevertheless, conventional descriptions fall short of structured information that effectively represents the interconnections among entities or attributes linked to a particular category. To address this limitation and prioritize harnessing structured knowledge, this paper advocates for leveraging LLMs to build a graph for each description to model the entities and attributes describing the category, as well as their correlations. Preexisting prompt tuning methods exhibit inadequacies in managing this structured knowledge. Consequently, we propose a novel approach called Hierarchical Prompt Tuning (HPT), which enables simultaneous modeling of both structured and conventional linguistic knowledge. Specifically, we introduce a relationship-guided attention module to capture pair-wise associations among entities and attributes for low-level prompt learning. In addition, by incorporating high-level and global-level prompts modeling overall semantics, the proposed hierarchical structure forges cross-level interlinks and empowers the model to handle more complex and long-term relationships. Extensive experiments demonstrate that our HPT shows strong effectiveness and generalizes much better than existing SOTA methods. Our code is available at https://github.com/Vill-Lab/2024-AAAI-HPT.

Graph Reasoning Transformers for Knowledge-Aware Question Answering

Augmenting Language Models (LMs) with structured knowledge graphs (KGs) aims to leverage structured world knowledge to enhance the capability of LMs to complete knowledge-intensive tasks. However, existing methods are unable to effectively utilize the structured knowledge in a KG due to their inability to capture the rich relational semantics of knowledge triplets. Moreover, the modality gap between natural language text and KGs has become a challenging obstacle when aligning and fusing cross-modal information. To address these challenges, we propose a novel knowledge-augmented question answering (QA) model, namely, Graph Reasoning Transformers (GRT). Different from conventional node-level methods, the GRT serves knowledge triplets as atomic knowledge and utilize a triplet-level graph encoder to capture triplet-level graph features. Furthermore, to alleviate the negative effect of the modality gap on joint reasoning, we propose a representation alignment pretraining to align the cross-modal representations and introduce a cross-modal information fusion module with attention bias to enable fine-grained information fusion. Extensive experiments conducted on three knowledge-intensive QA benchmarks show that the GRT outperforms the state-of-the-art KG-augmented QA systems, demonstrating the effectiveness and adaptation of our proposed model.

Wikiformer: Pre-training with Structured Information of Wikipedia for Ad-Hoc Retrieval

With the development of deep learning and natural language processing techniques, pre-trained language models have been widely used to solve information retrieval (IR) problems. Benefiting from the pre-training and fine-tuning paradigm, these models achieve state-of-the-art performance. In previous works, plain texts in Wikipedia have been widely used in the pre-training stage. However, the rich structured information in Wikipedia, such as the titles, abstracts, hierarchical heading (multi-level title) structure, relationship between articles, references, hyperlink structures, and the writing organizations, has not been fully explored. In this paper, we devise four pre-training objectives tailored for IR tasks based on the structured knowledge of Wikipedia. Compared to existing pre-training methods, our approach can better capture the semantic knowledge in the training corpus by leveraging the human-edited structured data from Wikipedia. Experimental results on multiple IR benchmark datasets show the superior performance of our model in both zero-shot and fine-tuning settings compared to existing strong retrieval baselines. Besides, experimental results in biomedical and legal domains demonstrate that our approach achieves better performance in vertical domains compared to previous models, especially in scenarios where long text similarity matching is needed. The code is available at https://github.com/oneal2000/Wikiformer.

CyberQ: Generating Questions and Answers for Cybersecurity Education Using Knowledge Graph-Augmented LLMs

Building a skilled cybersecurity workforce is paramount to building a safer digital world. However, the diverse skill set, constantly emerging vulnerabilities, and deployment of new cyber threats make learning cybersecurity challenging. Traditional education methods struggle to cope with cybersecurity's rapidly evolving landscape and keep students engaged and motivated. Different studies on students' behaviors show that an interactive mode of education by engaging through a question-answering system or dialoguing is one of the most effective learning methodologies. There is a strong need to create advanced AI-enabled education tools to promote interactive learning in cybersecurity. Unfortunately, there are no publicly available standard question-answer datasets to build such systems for students and novice learners to learn cybersecurity concepts, tools, and techniques. The education course material and online question banks are unstructured and need to be validated and updated by domain experts, which is tedious when done manually. In this paper, we propose CyberGen, a novel unification of large language models (LLMs) and knowledge graphs (KG) to generate the questions and answers for cybersecurity automatically. Augmenting the structured knowledge from knowledge graphs in prompts improves factual reasoning and reduces hallucinations in LLMs. We used the knowledge triples from cybersecurity knowledge graphs (AISecKG) to design prompts for ChatGPT and generate questions and answers using different prompting techniques. Our question-answer dataset, CyberQ, contains around 4k pairs of questions and answers. The domain expert manually evaluated the random samples for consistency and correctness. We train the generative model using the CyberQ dataset for question answering task.

Structure-CLIP: Towards Scene Graph Knowledge to Enhance Multi-Modal Structured Representations

Large-scale vision-language pre-training has achieved significant performance in multi-modal understanding and generation tasks. However, existing methods often perform poorly on image-text matching tasks that require structured representations, i.e., representations of objects, attributes, and relations. The models cannot make a distinction between "An astronaut rides a horse" and "A horse rides an astronaut". This is because they fail to fully leverage structured knowledge when learning multi-modal representations. In this paper, we present an end-to-end framework Structure-CLIP, which integrates Scene Graph Knowledge (SGK) to enhance multi-modal structured representations. Firstly, we use scene graphs to guide the construction of semantic negative examples, which results in an increased emphasis on learning structured representations. Moreover, a Knowledge-Enhance Encoder (KEE) is proposed to leverage SGK as input to further enhance structured representations. To verify the effectiveness of the proposed framework, we pre-train our model with the aforementioned approaches and conduct experiments on downstream tasks. Experimental results demonstrate that Structure-CLIP achieves state-of-the-art (SOTA) performance on VG-Attribution and VG-Relation datasets, with 12.5% and 4.1% ahead of the multi-modal SOTA model respectively. Meanwhile, the results on MSCOCO indicate that Structure-CLIP significantly enhances the structured representations while maintaining the ability of general representations. Our code is available at https://github.com/zjukg/Structure-CLIP.

Towards Human-like Learning from Relational Structured Data

Relational structured data is a way of representing knowledge using nodes and edges, while also capturing the meaning of that knowledge in a structured form that can be used for machine learning. Compared with vision and natural language data, relational structured data represents and manipulates structured knowledge, which can be beneficial for tasks that involve reasoning or inference. On the other hand, vision and NLP deal more with unstructured data (like images and text), and they often require different types of models and algorithms to extract useful information or features from the data. Human-like Learning develops methods that can harness relational structures and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. With Human-like Learning, the learning algorithm is efficient and can adapt to new or unseen situations, which is crucial in real-world applications where environments may change unpredictably. Moreover, the models are easier for humans to understand and interpret, which is important for transparency and trust in AI systems. In this talk, we present our recent attempts towards human-like learning from relational structured data.

A Study to Assess the effectiveness of Planned Teaching Program on Knowledge regarding Non Pharmacological Techniques for Pain Relief during Child Birth among Pregnant mothers in selected Community Area, Satna

Background: The experience of childbirth is a subjective and multidimensional issue and each woman passes through it in a different way. It is one of the most beautiful episodes in a mother’s life, associated with joy, happiness, and celebration. However, a delivery is also related to negative emotions: fear, anxiety, low sense of security, and the expectation of pain. Pain is both sensory and emotional. As a sensation, pain is interpreted by each individual is based on personal, cultural, and social experiences. Comfort measures that provide natural pain relief can be very effective during labor and childbirth. Birthing techniques such as hydrotherapy, hypno-birthing, patterned breathing, relaxation, and visualization can increase the production of endogenous endorphins that bind to receptors in the brain for pain relief. Objective: The present study aimed to assess the effectiveness of planned teaching program on knowledge regarding non pharmacological techniques for pain relief during child birth among pregnant mothers in selected community area, Satna. Methodology: Pre-experimental, one group pre-test post-test design was adopted for this study. 30 mothers were selected by using simple random sampling method for this study. A pre test was conducted by using structured knowledge questionnaire and on the same day the planned teaching program regarding non pharmacological techniques for pain relief during child birth was administered to them. After 7 days, post test was conducted with the same questionnaire for the same group of mothers to assess the effectiveness of planned teaching program. Result: The mean score in the post test is 22.8 and the mean score in the pre test is 9.8. The calculated value of t is 51.1 which is highly significant. The result confirms that the planned teaching program was effective significantly in improving the knowledge on non pharmacological techniques for pain relief during child birth among pregnant mothers.

Investigating the impact of structured knowledge feedback on collaborative academic writing

Investigating the impact of structured knowledge feedback on collaborative academic writing

A Study to Assess Knowledge regarding Scrub Typhus among adults at a selected community area, Kidangoor

A descriptive study was conducted to assess the level of knowledge regarding scrub typhus among adults at a selected community area, Kidangoor. The study was conducted in selected area of 3rd ward of Kidangoor Panchayath. The study sample consist of 30 adults. Samples are selected by using convenient sample technique. Structured knowledge questionnaire used to collect data. The result shows that among 30 samples 10(33.33%) had poor knowledge, 12(40%) had average level of knowledge and 8(26.67%) had good knowledge.

A Study to Assess the Effectiveness of Structured Teaching Programme on Knowledge regarding Warning Signs and Prevention of Reproductive Cancer among Female Non Health Professionals in a selected Hospital, Mangaluru

Female reproductive cancer is cancer of the organs related to female reproductive system. In order to identify these cancers in early stage women needs to have knowledge regarding warning signs of these cancers. Possessing knowledge helps to reduce mortality and enhances number of cure rate among women’s at risk of reproductive cancer and it is one of the prime responsibility of all health care personnel’s to disseminate information regarding warning signs and prevention of reproductive cancer. The present investigation was aimed to evaluate the effectiveness of structured teaching programme on knowledge regarding warning signs and prevention of reproductive cancer among female non health professionals working in selected hospital and to find out the association between the pre test knowledge score regarding warning signs and prevention of reproductive cancer and their selected demographic variables. The research design used in the study was pre – experimental one group pre-test post-test and non-probability purposive sampling technique was used to draw sample. The data was collected from 60 non health professionals using the structured knowledge questionnaire. Structured teaching programme was conducted and post-test was administered after 7 days using the same structured knowledge questionnaire to find out the effectiveness of structured teaching programme. The results revealed that the overall mean percentage knowledge score in the pre-test and post-test were 34% and 74.56% respectively. Enhancement mean percentage knowledge score was found to be 40.56%. The statistical paired ‘t’ test (t= 44.44) indicates that enhancement in the mean percentage knowledge score was found to be significant at 0.05 level for all the aspects under study. The association found to be significant between pre-test knowledge scores & selected socio demographic variables such as age, educational status, occupation, monthly income, and information regarding reproductive cancer at 0.05 levels (P< 0.05). There is no significant association and some demographical variable such as religion, marital status and family history of cancer, presence of children.

A Knowledge and Evaluation Model to Support the Conservation of Abandoned Historical Centres in Inner Areas

The planning of interventions aimed at preserving the built heritage of inner areas is a complex process due to the fragility of these contexts. It should stem from adequate knowledge to support the recognition of qualities, resources, and potentials, and the reinterpretation of residual values. From the perspective of an axiological approach to the built heritage, it is possible to foster the resemantization of such values based on a rigorous cognitive model. This research proposed a cognitive model of the built heritage of the historic neighbourhood of Granfonte in Leonforte (Enna). The knowledge model, developed in Excel, has a hierarchical type of structure characterized by domain, classes, properties, and the attribution of values to properties. The approach makes it possible to execute queries that arise from specific relationships between classes. In this study, we developed both simple queries to measure the percentages of certain characteristics of the building units and complex queries for the estimation of aggregate indices to define the degree of transformation and loss of integrity ITI and degradation ID of the building units or to identify those most exposed to the risk of ruination and contagion. The proposed model can be framed within the framework of ontologies supporting structured heritage knowledge.