Detecting and Measuring Depression on Social Media by Machine Learning Approach: a Review (Preprint)

BACKGROUND Detection of depression gained prominence soon after this troublesome disease emerged as a serious public health concern worldwide. OBJECTIVE The aim of this review intends to summarize methods to measure depressive symptoms on social media by machine learning approaches. METHODS A bibliographic retrieval was conducted from January 1990 until December 2020 in Google Scholar, PubMed, Medline, ERIC, PsycINFO, and BioMed. Seventeen studies met the inclusion criteria. RESULTS Of the fifteen studies, ten defined depression based on self-reported mental status, five defined based on self-declared mental status, and the last two were based on community membership. Besides, among the fifteen studies, thirteen conducted depression detection with Supervised Learning (SL) approaches, three used Unsupervised Learning (UL) approaches to detect depression, while the remaining one did not report the ML approach. Challenges such as sample scale, optimizing of predicting approaches and features, generalizability, issues about privacy, and ethic are still open to research. CONCLUSIONS ML approaches might work effectively for depression detection using text data from users on social media and it could serve as a complementary tool in practice about public psychological health.

Radiofrequency Technologies of Local Positioning in Healthcare

Introduction. Localization of objects position in closed space plays an important role in many areas of human activity, including medicine. Using indoor-positioning technologies as a part of telemedicine systems allows one to improve the quality of medical care and to reduce mortality of patients. Therefore, indoor-positioning technologies contribute to achieve the goals outlined in the Russian Federation government`s program "Healthcare development". Aim. To study the applicability of modern radiofrequency technologies for localization of patients inside a hospital building. Materials and methods. Scientific sources devoted to indoor-positioning based on radiofrequency technologies were analyzed. The methods used included: - bibliographic retrieval; - selection and verification of sources based on their relevance; - analysis of sources by methods of deconstruction and comparative analysis . Results. The result of the analysis indicated that radiofrequency positioning technologies allow one to locate objects using radio waves properties. The disadvantage of the technology is the penetration of radio signal through walls and floors. Given this, it is necessary to use complex algorithms to detect an object with accuracy to a specific room. Despite this disadvantage, radiofrequency technologies can be used for positioning in medical facilities since they are easy in deployment and service. Also, they are used in ready-made commercial solutions. ZigBee technology is an exception because it does not allow one to track moving objects in real-time. Conclusion. Based on the study it was concluded that BLE technology is the most suitable for indoor-positioning in medical facilities. It is energy-efficient, it has sufficiently fast data transfer rate, good communication radius and a large range of ready-made communication equipment. It is also worth noting that most wireless medical sensors exchange data via the BLE interface.

Circulating lncRNAs as noninvasive biomarkers in bladder cancer: A diagnostic meta-analysis based on 15 published articles

Objective Owing to inconsistency between reports, a meta-analysis was designed to appraise the clinical implications of long non-coding RNAs (lncRNAs) in urine and blood for the diagnosis of bladder cancer. Methods Studies that met the criteria were acquired by bibliographic retrieval through PubMed, Embase, and the Cochrane Library. The pooled diagnostic performance was evaluated by calculating the area under the summary receiver operator characteristic (SROC) curve. The potential sources of heterogeneity were approached through meta-regression and subgroup analyses. All statistical analyses and plots were performed by RevMan 5.3, Meta-DiSc 1.4, and STATA 12.0. Results A total of 43 studies from 15 articles consisting of 3370 bladder cancer patients and 3212 controls were incorporated in our meta-analysis. lncRNAs in urine and blood performed relatively well in diagnosing bladder cancer, with a pooled sensitivity of 0.78, a specificity of 0.79, and an area under the SROC curve (AUC) of 0.86. H19 displayed the best diagnostic accuracy with a pooled AUC of 0.90, followed by UCA1 and MALAT1. The heterogeneity among studies was partly conducted by sample size, lncRNA existence form (cell-free or intracellular lncRNA), lncRNA origin (exosome- or non-exosome-based lncRNA), lncRNA profiling (single- or multiple-lncRNA), specimen types, and ethnicity. Conclusions lncRNAs in urine and blood may serve as noninvasive diagnostic biomarkers with great promise for bladder cancer, while their clinical values need to be examined through further synthetic forward-looking studies.

A review on risk-taking in tournaments

Purpose This paper aims to summarize antecedents and consequences of risk-taking in tournaments and show the development of tournament optimization considering risk-taking. Moreover, further expansion expectations related to antecedents and consequences of risk-taking in tournaments and tournament optimization considering risk-taking are discussed. Design/methodology/approach A comprehensive bibliographic retrieval and further literature review and systematics re-organization are used to build the framework with respect to risk-taking in tournaments. Then, qualitative analysis is used to present conclusions of existing research. Findings By summarizing various antecedents, different consequences and tournament optimizations with regard to risk-taking in tournaments of existing research, the authors present a series of research opportunities regarding risk-taking in tournaments that can propel the advancement of tournament theory. Originality/value The studies on risk-taking in tournament have been recently received wide attention and are growing vigorously. Based on the summary and re-organization, the framework of literature studying risk-taking in tournaments is built. This literature review also helps researchers learn the advance of risk-taking in tournament and provides fruitful direction for future research on this topic.

Semantic information retrieval research based on co-occurrence analysis

Purpose – The paper aims to develop a new method for potential relations retrieval. It aims to find common aspects between co-occurrence analysis and ontology to build a model of semantic information retrieval based on co-occurrence analysis. Design/methodology/approach – This paper used a literature review, co-occurrence analysis, ontology build and other methods to design a model and process of semantic information retrieval based on co-occurrence analysis. Archaeological data from Wuhan University Library's bibliographic retrieval systems was used for experimental analysis. Findings – The literature review found that semantic information retrieval research mainly concentrates on ontology-based query techniques, semantic annotation and semantic relation retrieval. Moreover most recent systems can only achieve obvious relations retrieval. Ontology and co-occurrence analysis have strong similarities in theoretical ideas, data types, expressions, and applications. Research limitations/implications – The experiment data came from a Chinese university which perhaps limits its usefulness elsewhere. Practical implications – This paper constructed a model to understand potential relations retrieval. An experiment proved the feasibility of co-occurrence analysis used in semantic information retrieval. Compared with traditional retrieval, semantic information retrieval based on co-occurrence analysis is more user-friendly. Originality/value – This study is one of the first to combine co-occurrence analysis with semantic information retrieval to find detailed relationships.

Comparison of Contributing Factors in System Affective Response in a Boolean and Hypertext Bibliographic Database Environment

The present study compares factors of searcher affective response to two bibliographic retrieval systems. One system relied on hypertext linking to facilitate searching, the other a traditional Boolean approach. Eighty-five participants were randomly assigned to either system to perform four search tasks. Data collected from study participants were analyzed using factor analysis. . .