Applying Machine Learning and Data Fusion to the "Missing Person" Problem

Author(s):  
KMA Solaiman ◽  
Tao Sun ◽  
Alina Nesen ◽  
Bharat Bhargava ◽  
Michael Stonebraker

We present a system for integrating multiple sources of data for finding missing persons. This system can assist authorities in finding children during amber alerts, mentally challenged persons who have wandered off, or person-of-interests in an investigation. Authorities search for the person in question by reaching out to acquaintances, checking video feeds, or by looking into the previous histories relevant to the investigation. In the absence of any leads, authorities lean on public help from sources such as tweets or tip lines. A missing person investigation requires information from multiple modalities and heterogeneous data sources to be combined.<div>Existing cross-modal fusion models use separate information models for each data modality and lack the compatibility to utilize pre-existing object properties in an application domain. A framework for multimodal information retrieval, called Find-Them is developed. It includes extracting features from different modalities and mapping them into a standard schema for context-based data fusion. Find-Them can integrate application domains with previously derived object properties and can deliver data relevant for the mission objective based on the context and needs of the user. Measurements on a novel open-world cross-media dataset show the efficacy of our model. The objective of this work is to assist authorities in finding uses of Find-Them in missing person investigation.</div>

2021 ◽  
Author(s):  
KMA Solaiman ◽  
Tao Sun ◽  
Alina Nesen ◽  
Bharat Bhargava ◽  
Michael Stonebraker

We present a system for integrating multiple sources of data for finding missing persons. This system can assist authorities in finding children during amber alerts, mentally challenged persons who have wandered off, or person-of-interests in an investigation. Authorities search for the person in question by reaching out to acquaintances, checking video feeds, or by looking into the previous histories relevant to the investigation. In the absence of any leads, authorities lean on public help from sources such as tweets or tip lines. A missing person investigation requires information from multiple modalities and heterogeneous data sources to be combined.<div>Existing cross-modal fusion models use separate information models for each data modality and lack the compatibility to utilize pre-existing object properties in an application domain. A framework for multimodal information retrieval, called Find-Them is developed. It includes extracting features from different modalities and mapping them into a standard schema for context-based data fusion. Find-Them can integrate application domains with previously derived object properties and can deliver data relevant for the mission objective based on the context and needs of the user. Measurements on a novel open-world cross-media dataset show the efficacy of our model. The objective of this work is to assist authorities in finding uses of Find-Them in missing person investigation.</div>


2020 ◽  
Vol 10 (1) ◽  
pp. 7
Author(s):  
Miguel R. Luaces ◽  
Jesús A. Fisteus ◽  
Luis Sánchez-Fernández ◽  
Mario Munoz-Organero ◽  
Jesús Balado ◽  
...  

Providing citizens with the ability to move around in an accessible way is a requirement for all cities today. However, modeling city infrastructures so that accessible routes can be computed is a challenge because it involves collecting information from multiple, large-scale and heterogeneous data sources. In this paper, we propose and validate the architecture of an information system that creates an accessibility data model for cities by ingesting data from different types of sources and provides an application that can be used by people with different abilities to compute accessible routes. The article describes the processes that allow building a network of pedestrian infrastructures from the OpenStreetMap information (i.e., sidewalks and pedestrian crossings), improving the network with information extracted obtained from mobile-sensed LiDAR data (i.e., ramps, steps, and pedestrian crossings), detecting obstacles using volunteered information collected from the hardware sensors of the mobile devices of the citizens (i.e., ramps and steps), and detecting accessibility problems with software sensors in social networks (i.e., Twitter). The information system is validated through its application in a case study in the city of Vigo (Spain).


2012 ◽  
Vol 518-523 ◽  
pp. 1334-1339
Author(s):  
Jian Rang Zhang ◽  
Qing Tao Shen

In view of the complexity, redundancy and uncertainty of measuring data generated by mine environment monitoring systems, a structure of two level data fusion, an adaptive weighted first level fusion and a second level fusion of grey correlation analysis, is presented, thus to achieve the fusion for the monitoring data from heterogeneous data sources and the fusion for the data from heterogeneous sources. Application examples shows that the fusion model has stable performance with strong anti- interference and can be handled easily.


2021 ◽  
Vol 30 (1) ◽  
pp. 947-965
Author(s):  
Shafiza Ariffin Kashinath ◽  
Salama A. Mostafa ◽  
David Lim ◽  
Aida Mustapha ◽  
Hanayanti Hafit ◽  
...  

Abstract Designing a data-responsive system requires accurate input to ensure efficient results. The growth of technology in sensing methods and the needs of various kinds of data greatly impact data fusion (DF)-related study. A coordinative DF framework entails the participation of many subsystems or modules to produce coordinative features. These features are utilized to facilitate and improve solving certain domain problems. Consequently, this paper proposes a general Multiple Coordinative Data Fusion Modules (MCDFM) framework for real-time and heterogeneous data sources. We develop the MCDFM framework to adapt various DF application domains requiring macro and micro perspectives of the observed problems. This framework consists of preprocessing, filtering, and decision as key DF processing phases. These three phases integrate specific purpose algorithms or methods such as data cleaning and windowing methods for preprocessing, extended Kalman filter (EKF) for filtering, fuzzy logic for local decision, and software agents for coordinative decision. These methods perform tasks that assist in achieving local and coordinative decisions for each node in the network of the framework application domain. We illustrate and discuss the proposed framework in detail by taking a stretch of road intersections controlled by a traffic light controller (TLC) as a case study. The case study provides a clearer view of the way the proposed framework solves traffic congestion as a domain problem. We identify the traffic features that include the average vehicle count, average vehicle speed (km/h), average density (%), interval (s), and timestamp. The framework uses these features to identify three congestion periods, which are the nonpeak period with a congestion degree of 0.178 and a variance of 0.061, a medium peak period with a congestion degree of 0.588 and a variance of 0.0593, and a peak period with a congestion degree of 0.796 and a variance of 0.0296. The results of the TLC case study show that the framework provides various capabilities and flexibility features of both micro and macro views of the scenarios being observed and clearly presents viable solutions.


2007 ◽  
Vol 16 (01) ◽  
pp. 98-105
Author(s):  
V. Maojo ◽  
J. A. Mitchell ◽  
L. J. Frey

SummaryBiomedical Informatics as a whole faces a difficult epistemological task, since there is no foundation to explain the complexities of modeling clinical medicine and the many relationships between genotype, phenotype, and environment. This paper discusses current efforts to investigate such relationships, intended to lead to better diagnostic and therapeutic procedures and the development of treatments that could make personalized medicine a reality.To achieve this goal there are a number of issues to overcome. Primary are the rapidly growing numbers of heterogeneous data sources which must be integrated to support personalized medicine. Solutions involving the use of domain driven information models of heterogeneous data sources are described in conjunction with controlled ontologies and terminologies. A number of such applications are discussed.Researchers have realized that many dimensions of biology and medicine aim to understand and model the informational mechanisms that support more precise clinical diagnostic, prognostic and therapeutic procedures. As long as data grows exponentially, novel Biomedical Informatics approaches and tools are needed to manage the data. Although researchers are typically able to manage this information within specific, usually narrow contexts of clinical investigation, novel approaches for both training and clinical usage must be developed.After some preliminary overoptimistic expectations, it seems clear now that genetics alone cannot transform medicine. In order to achieve this, heterogeneous clinical and genomic data source must be integrated in scientifically meaningful and productive systems. This will include hypothesis-driven scientific research systems along with well understood information systems to support such research. These in turn will enable the faster advancement of personalized medicine.


2020 ◽  
Vol 12 (14) ◽  
pp. 5595 ◽  
Author(s):  
Ana Lavalle ◽  
Miguel A. Teruel ◽  
Alejandro Maté ◽  
Juan Trujillo

Fostering sustainability is paramount for Smart Cities development. Lately, Smart Cities are benefiting from the rising of Big Data coming from IoT devices, leading to improvements on monitoring and prevention. However, monitoring and prevention processes require visualization techniques as a key component. Indeed, in order to prevent possible hazards (such as fires, leaks, etc.) and optimize their resources, Smart Cities require adequate visualizations that provide insights to decision makers. Nevertheless, visualization of Big Data has always been a challenging issue, especially when such data are originated in real-time. This problem becomes even bigger in Smart City environments since we have to deal with many different groups of users and multiple heterogeneous data sources. Without a proper visualization methodology, complex dashboards including data from different nature are difficult to understand. In order to tackle this issue, we propose a methodology based on visualization techniques for Big Data, aimed at improving the evidence-gathering process by assisting users in the decision making in the context of Smart Cities. Moreover, in order to assess the impact of our proposal, a case study based on service calls for a fire department is presented. In this sense, our findings will be applied to data coming from citizen calls. Thus, the results of this work will contribute to the optimization of resources, namely fire extinguishing battalions, helping to improve their effectiveness and, as a result, the sustainability of a Smart City, operating better with less resources. Finally, in order to evaluate the impact of our proposal, we have performed an experiment, with non-expert users in data visualization.


Sign in / Sign up

Export Citation Format

Share Document