Michael B. Spring

Visualization of a document collection: the vibe system

Abstract The idea of using visualization for document retrieval is introduced through a new paradigm for query response handling. The paradigm is based on parallel queries or points of interest . Each point of interest is defined by a number of key terms and a display position. Documents, represented by icons, are positioned in the display based on the frequency count of word matches in the document to key terms in the points of interest. This visualization method has been implemented through a visualization system called VIBE.

Designing websites for persons with cognitive deficits: Design and usability of a psychoeducational intervention for persons with severe mental illness.

The purpose of this study was to develop an understanding of the design elements that influence the ability of persons with severe mental illness (SMI) and cognitive deficits to use a website, and to use this knowledge to design a web-based telehealth application to deliver a psychoeducation program to persons with schizophrenia and their families. Usability testing was conducted with 98 persons with SMI. First, individual website design elements were tested. Based on these results, theoretical website design models were used to create several alternative websites. These designs were tested for their ability to facilitate use by persons with SMI. The final website design is presented. The results indicate that commonly prescribed design models and guidelines produce websites that are poorly suited and confusing to persons with SMI. Our findings suggest an alternative model that should be considered when designing websites and other telehealth interventions for this population. Implications for future studies addressing the characteristics of accessible designs for persons with SMI and cognitive deficits are discussed.

An adaptive ontology mapping approach with neural network based constraint satisfaction

Ontology mapping seeks to find semantic correspondences between similar elements of different ontologies. It is a key challenge to achieve semantic interoperability in building the Semantic Web. This paper proposes a new generic and adaptive ontology mapping approach, called the PRIOR+, based on propagation theory, information retrieval techniques and artificial intelligence. The approach consists of three major modules, i.e., the IR-based similarity generator, the adaptive similarity filter and weighted similarity aggregator, and the neural network based constraint satisfaction solver. The approach first measures both linguistic and structural similarity of ontologies in a vector space model, and then aggregates them using an adaptive method based on their harmonies, which is defined as an estimator of performance of similarity. Finally to improve mapping accuracy the interactive activation and competition neural network is activated, if necessary, to search for a solution that can satisfy ontology constraints. The experimental results show that harmony is a good estimator of f-measure; the harmony based adaptive aggregation outperforms other aggregation methods; neural network approach significantly boosts the performance in most cases. Our approach is competitive with top-ranked systems on benchmark tests at OAEI campaign 2007, and performs the best on real cases in OAEI benchmark tests.

A Clinical Trial to Test the Feasibility of a Telehealth Psychoeducational Intervention for Persons With Schizophrenia and Their Families: Intervention and 3-Month Findings

OBJECTIVE To evaluate the feasibility of a telehealth psychoeducation intervention for persons with schizophrenia and their family members. STUDY DESIGN Randomized controlled trial. PARTICIPANTS 30 persons with schizophrenia and 21 family members or other informal support persons. INTERVENTIONS Web-based psychoeducation program that provided online group therapy and education. MAIN OUTCOME MEASURES Measures for persons with schizophrenia included perceived stress and perceived social support; for family members, they included disease-related distress and perceived social support. RESULTS At 3 months, participants with schizophrenia in the intervention group reported lower perceived stress (p = .04) and showed a trend for a higher perceived level of social support (p = .06). CONCLUSIONS The findings demonstrate the feasibility and impact of providing telehealth-based psychosocial treatments, including online therapy groups, to persons with schizophrenia and their families.

An interactive Web-based intervention for persons with TBI and their families: use and evaluation by female significant others.

ObjectiveTo assess the feasibility of providing in-home adjunctive and supportive services to persons with traumatic brain injury (TBI) and their families via a Web site. DesignNineteen families were provided with access to the Web site intervention for 6 months. Those who needed it were provided with a computer and Internet service in their homes. ParticipantsAdult women who were the significant others of adult males with moderate-to-severe TBI. Main Outcome MeasureValue and ease of use of the Web site. Each participants usage of the Web site was automatically tracked including each page visited, time of day, and time spent on the page. ResultsFemale significant others found the Web site to be valuable and easy to use, and used it throughout the 6-month period. The on-line support group was the most used and valued module. ConclusionsFamily caregivers will use Web-based interventions to help meet their needs for social support, information, and guidance following the return home of persons with TBI.

Critical Design Elements of E-Health Applications for Users With Severe Mental Illness: Singular Focus, Simple Architecture, Prominent Contents, Explicit Navigation, and Inclusive Hyperlinks

OBJECTIVE E-health applications are becoming integral components of general medical care delivery models and emerging for mental health care. Few exist for treatment of those with severe mental illness (SMI). In part, this is due to a lack of models to design such technologies for persons with cognitive impairments and lower technology experience. This study evaluated the effectiveness of an e-health design model for persons with SMI termed the Flat Explicit Design Model (FEDM). METHODS Persons with schizophrenia (n = 38) performed tasks to evaluate the effectiveness of 5 Web site designs: 4 were prominent public Web sites, and 1 was designed according to the FEDM. Linear mixed-effects regression models were used to examine differences in usability between the Web sites. Omnibus tests of between-site differences were conducted, followed by post hoc pairwise comparisons of means to examine specific Web site differences when omnibus tests reached statistical significance. RESULTS The Web site designed using the FEDM required less time to find information, had a higher success rate, and was rated easier to use and less frustrating than the other Web sites. The home page design of one of the other Web sites provided the best indication to users about a Web sites contents. The results are consistent with and were used to expand the FEDM. CONCLUSIONS The FEDM provides evidence-based guidelines to design e-health applications for person with SMI, including: minimize an applications layers or hierarchy, use explicit text, employ navigational memory aids, group hyperlinks in 1 area, and minimize the number of disparate subjects an application addresses.

Applying Social Annotations to Retrieve and Re-rank Web Resources

Web-base tagging systems, which include social bookmarking systems such as del.icio.us, have become increasingly popular. These systems allow participants to annotate or tag web resources. This paper examined the use of social annotations to improve the quality of web search. It involved two components. First, social annotations were used to index resources. Two annotation-based indexing methods were proposed. Second, social annotations were used to improve search result ranking. Four annotation-based ranking methods were proposed. The result showed that using only annotation as an index of resources may not be appropriate. Since social annotations could be viewed as a high level concept of the content, combining them to the content of resource could add some more important concepts to the resources. The result also suggested that both static feature and similarity feature should be considered when using social annotations to re-rank search result.

Ontology Mapping: As a Binary Classification Problem

Ontology mapping seeks to find semantic correspondences between similar elements of different ontologies. Ontology mapping is critical to achieve semantic interoperability in the WWW. To solve the ontology mapping problem, this paper proposes a non-instance learning-based approach that transforms the ontology mapping problem to a binary classification problem and utilizes machine learning techniques as a solution. Same as other machine learning based approaches, a number of features (i.e., linguistic, structural and web features)are generated for each mapping candidate. However, in contrast to other learning-based mapping approaches, the features proposed in our approach are generic and do not rely on the existence and sufficiency of instances. Therefore our approach can be generalized to different domains without extra training efforts. To evaluate our approach, two experiments (i.e., within-task vs. cross-task) are implemented and the SVM algorithm is applied.Experimental results show that our non-instance learning-based ontology mapping approach performs well on most of OAEI benchmark tests when training and testing on the same mapping task; and the results of approach vary according to the likelihood of training data and testing data when training and testing on different mapping tasks.

The structure of IT standardization

■ We address the question of how best to develop standards for the information technology industry. While it may be possible to generalize our conclusions for other industry groups, the focus here is on the issues faced by the IT industries. arvin Minsky said “Anything that you hear about computers and AI should be ignored, because we’re in the Dark Ages. We’re in the thousand years between no technology and all technology.’’ [Brand 1988, p. 104] While information technology is probably in somewhat better shape than Minsky suggests is the case for AI, it is true that we are in a period of very rapid, and often unpredictable change. Bibliophiles will be familiar with the term “incunabula” as it refers to books produced between 1450 and 1500. The term more generally describes any art or industry in the early stages of development, and information technology can certainly be so described. Developments are both rapid and pervasive, and there are many indications of the rate and extent of change:

Randomized controlled trial of a collaborative care intervention to manage cancer-related symptoms: lessons learned

Background Collaborative care interventions to treat depression have begun to be tested in settings outside of primary care. However, few studies have expanded the collaborative care model to other settings and targeted comorbid physical symptoms of depression. Purpose The aims of this report were to: (1) describe the design and methods of a trial testing the efficacy of a stepped collaborative care intervention designed to manage cancer-related symptoms and improve overall quality of life in patients diagnosed with hepatobiliary carcinoma; and (2) share the lessons learned during the design, implementation, and evaluation of the trial. Methods The trial was a phase III randomized controlled trial testing the efficacy of a stepped collaborative care intervention to reduce depression, pain, and fatigue in patients diagnosed with advanced cancer. The intervention was compared to an enhanced usual care arm. The primary outcomes included the Center for Epidemiological Studies-Depression scale, Brief Pain Inventory, and Functional Assessment of Cancer Therapy (FACT)-Fatigue, and the FACT-Hepatobiliary. Sociodemographic and disease-specific characteristics were recorded from the medical record; Natural Killer cells and cytokines that are associated with these symptoms and with disease progression were assayed from serum. Results  and  Discussion The issues addressed include: (1) development of collaborative care in the context of oncology (e.g., timing of the intervention, tailoring of the intervention, ethical issues regarding randomization of patients, and changes in medical treatment over the course of the study); (2) use of a website by chronically ill populations (e.g., design and access to the website, development of the website and intervention, ethical issues associated with website development, website usage, and unanticipated costs associated with website development); (3) evaluation of the efficacy of intervention (e.g., patient preferences, proxy raters, changes in medical treatment, and inclusion of biomarkers as endpoints); and (4) analyses and interpretation of the intervention (e.g., confounding factors, dose and active ingredients, and risks and benefits of collaborative care interventions in chronically ill patients). Limitations The limitations to the study, although not fully realized at this time as the trial is ongoing, include: (1) heterogeneity of the diagnoses and treatments of participants; and (2) inclusion of caregivers as proxy raters but not as participants in the intervention. Conclusions Collaborative care interventions to manage multiple symptoms in a tertiary cancer center are feasible. However, researchers designing and implementing interventions that are web-based, target multiple symptoms, and for oncology patients may benefit from previous experiences.