Contextually Relevant Research Articles

Application of Generative Artificial Intelligence Models for Accurate Prescription Label Identification and Information Retrieval for the Elderly in Northern East of Thailand.

This study introduces a novel AI-driven approach to support elderly patients in Thailand with medication management, focusing on accurate drug label interpretation. Two model architectures were explored: a Two-Stage Optical Character Recognition (OCR) and Large Language Model (LLM) pipeline combining EasyOCR with Qwen2-72b-instruct and a Uni-Stage Visual Question Answering (VQA) model using Qwen2-72b-VL. Both models operated in a zero-shot capacity, utilizing Retrieval-Augmented Generation (RAG) with DrugBank references to ensure contextual relevance and accuracy. Performance was evaluated on a dataset of 100 diverse prescription labels from Thai healthcare facilities, using RAG Assessment (RAGAs) metrics to assess Context Recall, Factual Correctness, Faithfulness, and Semantic Similarity. The Two-Stage model achieved high accuracy (94%) and strong RAGAs scores, particularly in Context Recall (0.88) and Semantic Similarity (0.91), making it well-suited for complex medication instructions. In contrast, the Uni-Stage model delivered faster response times, making it practical for high-volume environments such as pharmacies. This study demonstrates the potential of zero-shot AI models in addressing medication management challenges for the elderly by providing clear, accurate, and contextually relevant label interpretations. The findings underscore the adaptability of AI in healthcare, balancing accuracy and efficiency to meet various real-world needs.

Development and Evaluation of a Web-Based Platform for Personalized Educational and Professional Assistance for Dementia Caregivers: Proposal for a Mixed Methods Study.

Alzheimer disease (AD) and AD-related dementia are prevalent concerns for aging populations. With a growing older adult population living in the United States, the number of people living with dementia is expected to grow, posing significant challenges for informal caregivers. The mental and physical burdens associated with caregiving highlight the importance of developing novel and effective resources to support caregivers. However, technology solutions designed to address their needs often face low adoption rates due to usability issues and a lack of contextual relevance. This study focuses on developing a web-based platform providing financial and legal planning information and education for dementia caregivers and evaluating the platform's usability and adoptability. The goal of this project is to create a web-based platform that connects caregivers with personalized and easily accessible resources. This project involves industrial, academic, and community partners and focuses on two primary aims: (1) developing a digital platform using a Dementia Care Personalization Algorithm and assessing feasibility in a pilot group of caregivers, and (2) evaluating the acceptability and usability of the digital platform across different racial or ethnic populations. This work will aid in the development of technology-based interventions to reduce caregiver burden. The phase I study follows an iterative Design Thinking approach, involving at least 25 dementia caregivers as a user feedback panel to assess the platform's functionality, aesthetics, information, and overall quality using the adapted Mobile Application Rating Scale. Phase II is a usability study with 300 dementia caregivers in Texas (100 African American, 100 Hispanic or Latinx, and 100 non-Hispanic White). Participants will use the digital platform for about 4 weeks and evaluate its usefulness and ease of use through the Technology Acceptance Survey. The study received funding from the National Institute on Aging on September 3, 2021. Ethical approval for phase I was obtained from the Texas A&M University Institutional Review Board on December 8, 2021, with data collection starting on January 1, 2022, and concluding on May 31, 2022. Phase I results were published on September 5, 2023, and April 17, 2024, respectively. On June 21, 2023, ethical approval for human subjects for phase II was granted, and participant recruitment began on July 1, 2023. Upon completing these aims, we expect to deliver a widely accessible digital platform tailored to assist dementia caregivers with financial and legal challenges by connecting them to personalized, contextually relevant information and resources in Texas. If successful, we plan to work with caregiving organizations to scale and sustain the platform, addressing the needs of the growing population living with dementia. DERR1-10.2196/64127.

Different mechanisms of contextual inference govern associatively learned and sensory-evoked postural responses

Human standing balance relies on the continuous monitoring and integration of sensory signals to infer our body's motion and orientation within the environment. However, when sensory information is no longer contextually relevant to balancing the body (e.g., when sensory and motor signals are incongruent), sensory-evoked balance responses are rapidly suppressed, much earlier than any conscious perception of changes in balance control. Here, we used a robotic balance simulator to assess whether associatively learned postural responses are similarly modulated by sensorimotor incongruence and contextual relevance to postural control. Twenty-nine participants in three groups were classically conditioned to generate postural responses to whole-body perturbations when presented with an initially neutral sound cue. During catch and extinction trials, participants received only the auditory stimulus but in different sensorimotor states corresponding to their group: 1) during normal active balance, 2) while immobilized, and 3) throughout periods where the computer subtly removed active control over balance. In the balancing and immobilized states, conditioned responses were either evoked or suppressed, respectively, according to the (in)ability to control movement. Following the immobilized state, conditioned responses were renewed when balance was restored, indicating that conditioning was retained but only expressed when contextually relevant. In contrast, conditioned responses persisted in the computer-controlled state even though there was no causal relationship between motor and sensory signals. These findings suggest that mechanisms responsible for sensory-evoked and conditioned postural responses do not share a single, central contextual inference and assessment of their relevance to postural control, and may instead operate in parallel.

Enhancing Image Captioning Using Deep Convolutional Generative Adversarial Networks

Introduction:: Introduction: Image caption generation has long been a fundamental challenge in the area of computer vision (CV) and natural language processing (NLP). In this research, we present an innovative approach that harnesses the power of Deep Convolutional Generative Adversarial Networks (DCGAN) and adversarial training to revolutionize the generation of natural and contextually relevant image captions. Method:: Our method significantly improves the fluency, coherence, and contextual relevance of generated captions and showcases the effectiveness of RL reward-based fine-tuning. Through a comprehensive evaluation of COCO datasets, our model demonstrates superior performance over baseline and state-of-the-art methods. On the COCO dataset, our model outperforms current state-of-the-art (SOTA) models across all metrics, achieving BLEU-4 (0.327), METEOR (0.249), Rough (0.525) and CIDEr (1.155) scores. Result:: The integration of DCGAN and adversarial training opens new possibilities in image captioning, with applications spanning from automated content generation to enhanced accessibility solutions. Conclusion:: This research paves the way for more intelligent and context-aware image understanding systems, promising exciting future exploration and innovation prospects.

Image Captioning of an Environment Using Machine Learning Algorithms (A Case Study of Gwarzo Road, Kano Nigeria)

This paper investigates the application of machine learning algorithms for automatic image captioning, focusing on a case study of Gwarzo Road in Kano, Nigeria. The research aims to design a robust VGG16/LSTM-based model that generates accurate and contextually relevant descriptions for images captured along the Kabuga to Bayero University Kano new site route. The methodology involves collecting images at three distinct times of the day (morning, afternoon, and evening) over 60 days, resizing and labelling them with relevant captions to build a comprehensive dataset. The VGG16 model, known for its efficiency in image processing, was employed for feature extraction, while the LSTM network was used to generate captions by interpreting the contextual and semantic details of the images. This study addresses key challenges in image captioning, such as localized object detection and generating meaningful textual descriptions, improving on existing datasets and models that often lack contextual relevance in specific environments. The expected outcomes of this research include the development of a precise caption generation model with high accuracy and efficiency. The resulting model achieved a BLEU score of 0.051, representing baseline performance in caption generation with partial alignment to human-generated references. Additionally, the model’s highest accuracy based on the loss function reached 55%, while the lowest accuracy was 50%, with an average accuracy of 53%. The creation of a localized image database further enhances the significance of this research for future applications and studies in image captioning.

GastroBot: a Chinese gastrointestinal disease chatbot based on the retrieval-augmented generation.

Large Language Models (LLMs) play a crucial role in clinical information processing, showcasing robust generalization across diverse language tasks. However, existing LLMs, despite their significance, lack optimization for clinical applications, presenting challenges in terms of illusions and interpretability. The Retrieval-Augmented Generation (RAG) model addresses these issues by providing sources for answer generation, thereby reducing errors. This study explores the application of RAG technology in clinical gastroenterology to enhance knowledge generation on gastrointestinal diseases. We fine-tuned the embedding model using a corpus consisting of 25 guidelines on gastrointestinal diseases. The fine-tuned model exhibited an 18% improvement in hit rate compared to its base model, gte-base-zh. Moreover, it outperformed OpenAI's Embedding model by 20%. Employing the RAG framework with the llama-index, we developed a Chinese gastroenterology chatbot named "GastroBot," which significantly improves answer accuracy and contextual relevance, minimizing errors and the risk of disseminating misleading information. When evaluating GastroBot using the RAGAS framework, we observed a context recall rate of 95%. The faithfulness to the source, stands at 93.73%. The relevance of answers exhibits a strong correlation, reaching 92.28%. These findings highlight the effectiveness of GastroBot in providing accurate and contextually relevant information about gastrointestinal diseases. During manual assessment of GastroBot, in comparison with other models, our GastroBot model delivers a substantial amount of valuable knowledge while ensuring the completeness and consistency of the results. Research findings suggest that incorporating the RAG method into clinical gastroenterology can enhance the accuracy and reliability of large language models. Serving as a practical implementation of this method, GastroBot has demonstrated significant enhancements in contextual comprehension and response quality. Continued exploration and refinement of the model are poised to drive forward clinical information processing and decision support in the gastroenterology field.

Patterns of presentation of adults with hearing impairment in a peri-urban community in South Africa: a qualitative study

BackgroundThere is a wealth of research, globally, on the help-seeking behavior patterns of individuals with a variety of chronic medical conditions. Contextually relevant findings have been reported describing the influence of a disease and/or disorder together with the related personal, social, environmental, and region-specific factors. However, research related to help-seeking behavior patterns in hearing impairment is mostly found in high income countries (HICs) and tells a one-sided story indicating a knowledge gap in other contexts. As part of a bigger study titled “In pursuit of preventive audiology: Help-seeking behavior patterns of adults with hearing impairment in a peri-urban community in South Africa”, the main aim of this study was to describe the patterns of presentation of symptoms in adults with hearing impairment from a peri-urban community in South Africa.MethodsThrough a non-probability purposive sampling method, 23 adults with hearing impairment participated in semi-structured interviews using an interview guide from April – June 2021. Thematic analysis through a deductive analysis approach was adopted for data analysis.ResultsParticipants’ patterns of presentation indicated a belief in a Western bio-medical causation to explain their hearing impairment. Help-seekers sought help mostly from healthcare workers at all levels of healthcare in the public (primary, secondary, and tertiary) and private healthcare sectors, followed by a few participants consulting their social networks. Only one participant used a traditional healthcare provider to resolve his hearing difficulties.ConclusionsIn a diverse country like South Africa, with a majority African population, the influence of cultural practices and beliefs proved to have a minimal influence on participants’ help-seeking behavior for their ear and hearing challenges. However, an interplay of factors related to one’s context at the time played a significant role in the patterns of presentation to the ear and hearing clinic. There is therefore a need to understand the perspectives of adults with hearing impairment from their environments to facilitate more contextual relevancy in healthcare provision within the preventive audiology field. Hearing health policymakers should also endeavor to consider the realities of health-seeking in each context and environment.

Is Organizational Evaluation Competency Framework Contextually Relevant for Local NGO Strategy Measurement? Evidence from Uganda

Despite the assumed added value of localization, what remains unclear is the method and framework within which local non-governmental organization (NGO) strategies are measured. Owing to considerable difference between donor demands and local stakeholder’s requirements, local NGO evaluation landscape in many developing countries including Uganda is characterized by disrupted systematic application of evaluation practices and standards. While there are considerable congregation towards ensuring standard evaluation, various evaluation models and frameworks; the proliferation of different donor donors’ framework is increasingly making strategy evaluation as subject of varied imagination against localized norms and standards. This study sought to validate the contextual relevance of organizational evaluation competency (OEC) framework as a standard process upon which local NGOs could meaningfully be seen to contribute to the realization of national development plan and aspirations. In-depth key informant interviews were conducted with 16 respondents drawn from NGO executives, NGO managers, board members, consultants, and academicians from a wider spectrum of specializations. Findings revealed that the OEC framework is contextually relevant for strategy evaluation. Specifically, the findings revealed that matured NGOs strategies were positively impacted by utilization OEC framework as they consistently apply it compared to emerging new local NGO. The study highlight higher potentials in wider application of OEC framework by both matured and emerging NGOs. This implies policy engagement on capacity strengthening must urgently prioritize NGO network support towards a strengthened evaluation technical expertise and steering. Keywords: Localization, Strategy, Organizational evaluation competency framework, NGOs.

Why Doesn’t Executive Function Training Improve Academic Achievement? Rethinking Individual Differences, Relevance, and Engagement from a Contextual Framework

ABSTRACT Performance on lab assessments of executive functions predicts academic achievement and other positive life outcomes. A primary goal of research on executive functions has been to design interventions that improve outcomes like academic achievement by improving executive functions. These interventions typically involve extensive practice on abstract lab-based tasks and lead to improvements on these practiced tasks. However, interventions rarely improve performance on non-practiced tasks and rarely benefit outcomes like academic achievement. Contemporary frameworks of executive function development suggest that executive functions develop and are engaged within personal, social, historical, and cultural contexts. lab-based tasks do not well-capture the real-world contexts that require executive functions and should not be expected to provide generalized benefits outside of the lab. We propose a perspective for understanding individual differences in performance on executive function assessments that focuses on contextual influences on executive functions. We extend this contextual approach to training executive function engagement, rather than training executive functions directly. First, interventions should incorporate task content that is contextually relevant to the targeted outcome. Second, interventions should encourage engaging executive functions through reinforcement and contextual relevance, which may better translate to real-world outcomes than training executive functions directly. While such individualized executive functions interventions do not address systemic factors that greatly impact outcomes like academic achievement, given the extensive resources devoted to improving executive functions, we hypothesize that interventions designed to encourage children’s engagement of executive functions hold more promise for impacting real-world outcomes than interventions designed to improve executive function capacities.

A South-to-South Cultural Adaptation of an Evidence-Based Parenting Program for Families in the Philippines.

Rates of child maltreatment are higher in low‐ and middle‐income countries due to risk factors such as social inequities, economic adversity, and sociocultural norms. Given the evidence showing the effectiveness of parenting interventions to prevent child maltreatment, this study embarked on a cultural adaptation of an evidence‐based parenting program with the eventual goal of integrating it within a nationwide conditional cash transfer program for low‐income Filipino parents with children aged 2–6 years. We document the systematic adaptation of the Parenting for Lifelong Health for Young Children program that was developed and tested in South Africa, for low‐resource Filipino families using the heuristic framework for the cultural adaptation of interventions. We underscore the merits of conducting a multistage top‐down and bottom‐up process that uses a participatory approach among cultural insiders and outsiders to develop a parenting intervention that reflects the contextual realities and cultural values of end users. The adapted program, Masayang Pamilya Para sa Batang Pilipino, is the product of a delicate and deliberate effort to balance Filipino childrearing goals and values with the scientific evidence on components of parenting interventions known to promote positive parenting and prevent child maltreatment.

Advocating for Front-line Health Advocates- A Scoping Review of Health Advocacy Pedagogy

Abstract Background Health advocacy (HA) is a core competency of transformed, socially- and contextually-responsive, 21st century medical graduates though remains challenging for medical educators, including in South Africa. The optimal delivery of HA education has been vulnerable to vague definitions, lacking distinct or measurable dimensions, varied judgement, and poor prioritization in curricula which balance competing interests. We aimed to clarify HA definitions and explore current practices of HA pedagogy. Methods A step wise, scoping review approach was adopted, for evidence on HA and HA pedagogy. This included electronic database searches of published literature, and internet searches for non-journal publications, from 2000 to 2020 and limited to reports published in English. Websites of relevant regulatory bodies and professional organizations were also explored. Inductive thematic analysis was used to analyze the data. Results Sixty-four total articles and documents were included. Majority emerged from North America, the field of health education, and medical disciplines of family medicine, internal medicine, and pediatrics. HA compels graduates to be change agents and activists, at both the individual and community levels, and to promote equitable access to quality healthcare. It is increasingly associated with social accountability, consideration of the social determinants of health, and progression from advocating with, rather than for, individuals and populations. Various curricular approaches were evident, including didactic and experiential learning, and longitudinal and integrated designs. Conclusions Although HA is a core competency for medical graduates in South Africa, there is little evidence related to HA pedagogy in the country and the African continent. Despite distinct variability in current instructional approaches, there is potential for consolidation, contextual relevance, enhanced HA pedagogy, and improved HA competencies. Key messages HA is a core medical graduates’ competency but HA education is often sub-optimally engaged with and delivered. Enhanced, contextually relevant HA pedagogy could empower medical graduates to be better health advocates and more confident change agents for patients and communities they serve.

Methodological rigour and coherence in the construction of instruments: The emotional social screening tool for school readiness

The need for a contextually relevant and empirically grounded measure of emotional social competence in Grade R children was identified in the literature. The aim of this study was to develop a contextually relevant instrument for emotional social competence in preschool children. The study adopted a four-phase approach with each phase using distinct methodological approaches. This article reports on the use of multiple research methods to achieve methodological rigour and coherence in the construction. Phase 1 used systematic review methodology to establish a theoretical foundation for the instrument. The results identified two domains and nine subdomains that formed the theoretical model for the instrument. In phase 2 stakeholder perceptions of emotional and social competence were identified through concept mapping to increase contextual relevance and sensitivity. The results highlighted that early stimulation and contextual factors impacted school readiness and needed to be included. The construction of the instrument incorporated the findings from the first two stages. The draft instrument was presented to a panel of experts, using the Delphi technique, for validation of content and scalar decisions in phase 3. The results supported the proposed format and content of the screening tool. The resulting instrument was piloted in phase 4 with survey research. Good internal consistency was reported and the factor structure supported. The multiphase methodology provided an overarching framework with methodological rigour and coherence. The grounding in the literature, stakeholder consultation and rigorous validation processes enhanced the resultant instrument. The articulation of one phase into the next ensured methodological coherence.

Addressing curriculum decolonisation and education for sustainable development through epistemically diverse curricula

Abstract Transformation in Higher Education has been an ongoing concern in post-apartheid South Africa, especially in light of universities’ expected contribution to economic and socio-political transformation. In particular, curriculum transformation has proved challenging, as evidenced in actions and calls by students in recent years for decolonisation of the curriculum. This study, which formed part of an institutional response to the challenge of curriculum transformation and decolonisation, initially sought to examine perceptions of the term ‘decolonisation’ amongst a group of early career lecturers at a leading university in South Africa. Highlighted in the outcomes of the study was the centrality of personal and contextual relevance in notions of decolonised curricula, the impact of curriculum conversations on lecturers’ well-being, and the broader implications of responsive and relevant curricula for institutional and societal well-being. In this respect, the findings of the study illustrated the similarities of curriculum decolonisation approaches and the concept of education for sustainable development which is underpinned by the goal of global well-being and the common good. Also highlighted was the need for greater balance between Mode 1 (theoretical) and Mode 2 (contextually relevant) knowledge in curricula, leading us to posit that both curriculum decolonisation and education for sustainable development are equally necessary for institutional and broader societal reform and well-being, and that both imperatives may potentially be achieved by focusing on the principles of epistemically diverse curricula.

Adaptation and Validation of the Shame Questionnaire Among Ugandan Youth Living with HIV.

The purpose of this study was to adapt and validate a measure of HIV-related shame, the Shame Questionnaire (SQ), among Ugandan youth living with HIV. Culturally relevant, reliable and valid measurement is critical in the accurate assessment of HIV-related shame (a painful internalized emotion encompassing feelings that the self is damaged and defective) on psychosocial functioning, as well as the determination of the efficacy of interventions among youth living with HIV in sub-Saharan Africa. We utilized qualitative (i.e., cognitive interviews; N = 31) and quantitative (i.e., classical test theory and item response theory; N = 150) methods to establish, content, criterion and construct validity of the SQ. Cognitive interviews resulted in the revision in the wording of 2 out of 8 SQ items. Participants who endorsed having shame had statistically significant higher SQ scores than participants who did not endorse having shame (p < 0.001), suggesting criterion validity. We found a statistically significant positive relationship between SQ scores and average trauma symptom scores among participants (p < 0.001), also suggesting criterion validity. Finally, we found construct validity with discrimination parameters of the graded response IRT model all in the high range with a wide range of difficulty parameters across the 8 items of the SQ. Overall our results suggest that the SQ is a contextually relevant, valid and reliable assessment tool among Ugandan youth living with HIV. Findings provide support for the utilization of qualitative and quantitative methods in the adaptation of measures for cross-cultural use in order to maintain validity and contextual relevance.

Making sense of leadership in early childhood education: Tensions and complexities between concepts and practices

AbstractEffective leadership within early childhood settings is aligned with the perceived successful implementation of high quality care and education programmes (Thornton, Tamati, Clarkin-Philips, Aitken &amp; Wansbrough, 2009). With growing attention on the role early childhood education (ECE) plays in preparing children to be successful in their lives, it is not surprising that there is increased focus on the work and impact of educational leaders in this endeavour. An expanding body of research specifically exploring leadership within ECE settings illustrates how much of the educational leadership theory corpus lacks contextual relevance and fails to recognise the complexities and realities of leading in early years contexts (Rodd, 2013). The qualitative research reported on in this article examines the leadership understandings and perceptions of five qualified, registered early childhood leaders. The findings illustrated that whilst participants were cognisant of the role effective leadership plays in providing quality care and education, there was much confusion about what leadership entailed and how leadership differentiated from management in this context. The contextual complexities of the ECE sector were a significant influence on each participant’s opportunity to learn about, and practise leadership. Findings also revealed a need for contextually relevant and progressive approaches to leadership learning to support early childhood leaders and teachers in their leadership work.

Context preserving navigation redesign under Markovian assumption for responsive websites

Identification and sorting of contextually relevant links are important for navigation design of responsive websites. While average click ratio is often used as a metric to evaluate contextual relevance of navigation structure from users’ view, we propose a metric called implied deviation to quantify the same from designer point of view. Average click ratio minimization problem typically solved using meta-heuristics has issues such as loss of designer-defined contextual relevance and loss of connectivity among webpages. To solve this problem, we propose a deviation minimization framework to suggest context-preserving navigation structure. The proposed framework consists of three stages. Stage 1 models user navigation behavior as a Markov process and generates a transition probability matrix. Then we use the transition probabilities as weights to relax the original average click ratio minimization problem, and bring it to a form similar to a transportation model. The corresponding solution is considered as the initial basic feasible solution of the original problem. In Stage 2, transition probability guided meta-heuristics improve upon the initial basic feasible solution. Specifically, we use modified simulated annealing and ant colony optimization algorithms. This resulting solution is further weighted with transition probabilities to make a tradeoff in Stage 3. We obtained experimental results based on two education and university-related datasets to show that the proposed framework can achieve a good tradeoff for contextual relevance in terms of both the average click ratio and implied deviation.

131. Mapping the Views of Adolescent Health Stakeholders

131. Mapping the Views of Adolescent Health Stakeholders

Contextual Internet Multimedia Advertising

The advent of media-sharing sites has led to the unprecedented Internet delivery of community-contributed media like images and videos. Those visual contents have become the primary sources for online advertising. Conventional advertising treats multimedia advertising as general text advertising by displaying advertisements either relevant to the queries or the Web pages, without considering the potential advantages which could be brought by media contents. In this paper, we summarize the trend of Internet multimedia advertising and conduct a broad survey on the methodologies for advertising which are driven by the rich contents of images and videos. We discuss three key problems in a generic multimedia advertising framework. These problems are: contextual relevance that determines the selection of relevant advertisements, contextual intrusiveness which is the key to detect appropriate ad insertion positions within an image or video, and insertion optimization that achieves the best association between the advertisements and insertion positions so that the effectiveness of advertising can be maximized in terms of both contextual relevance and contextual intrusiveness. We show recently developed MediaSense which consists of image, video, and game advertising as an exemplary application of contextual multimedia advertising. In the MediaSense, the most contextually relevant ads are embedded at the most appropriate positions within images or videos. To this end, techniques in computer vision, multimedia retrieval, and computer human interaction are leveraged. We also envision that the next trend of multimedia advertising would be game-like advertising which is more impressionative and thus can promote advertising in an interactive, as well as more compelling and effective way. We conclude this survey with a brief outlook on open research directions.

VideoSense: A Contextual In-Video Advertising System

With Internet delivery of video content surging to an unprecedented level, video has become one of the primary sources for online advertising. In this paper, we present VideoSense as a novel contextual in-video advertising system, which automatically associates the relevant video ads and seamlessly inserts the ads at the appropriate positions within each individual video. Unlike most video sites which treat video advertising as general text advertising by displaying video ads at the beginning or the end of a video or around a video, VideoSense aims to embed more contextually relevant ads at less intrusive positions within the video stream. Specifically, given a Web page containing an online video, VideoSense is able to extract the surrounding text related to this video, detect a set of candidate ad insertion positions based on video content discontinuity and attractiveness, select a list of relevant candidate ads according to multimodal relevance. To support contextual advertising, we formulate this task as a nonlinear 0-1 integer programming problem by maximizing contextual relevance while minimizing content intrusiveness at the same time. The experiments proved the effectiveness of VideoSense for online video service.

Family-Based HIV Preventive Intervention

Summary Social indicators suggest that African American adolescents are in the highest risk categories of those contracting HIV/AIDS (CDC, 2001). The dramatic impact of HIV/AIDS on urban African American youth have influenced community leaders and policy makers to place high priority on programming that can prevent youth's exposure to the virus (Pequegnat & Szapocznik, 2000). Program developers are encouraged to design programs that reflect the developmental ecology of urban youth (Tolan, Gorman-Smith, & Henry, 2003). This often translates into three concrete programmatic features: (1) Contextual relevance; (2) Developmental-groundedness; and (3) Systemic Delivery. Because families are considered to be urban youth's best hope to grow up and survive multiple dangers in urban neighborhoods (Pequegnat& Szapocznik, 2000), centering prevention within families may ensure that youth receive ongoing support, education, and messages that can increase their capacity to negotiate peer situations involving sex. This paper will present preliminary data from an HIV/AIDS prevention program that is contextually relevant, developmentally grounded and systematically-delivered. The collaborative HIV/AIDS Adolescent Mental Health Project (CHAMP) is aimed at decreasing HIV/AIDS risk exposure among a sample of African American youth living in a poverty-stricken, inner-city community in Chicago. This study describes results from this family-based HIV preventive intervention and involves 88 African American pre-adolescents and their primary caregivers. We present results for the intervention group at baseline and post intervention. We compare post test results to a community comparison group of youth. Suggestions for future research are provided.