The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Artificially synthesized sentences are used for malicious purposes such as unsolicited commercial junk submission to Web site. We study a problem to distinguish between natural and synthesized messages generated in Markov chains and show experimental results. Based on the difficulties of the problem, we consider a new application to CAPTCHA, a type of challenge-response test used in computing to ensure...
This paper proposes a novel system for automatically detecting children from a color monocular back-up camera, as part of a back-up warning device in passenger vehicles. We presented the use of an attentional mechansim that focuses compute-intensive bounding-box classifiers on a subset of all possible bounding-box solutions to enable real-time performance of 248ms per frame with negligible reduction...
We propose a new method for audio-visual sensor fusion and apply it to automatic aggression detection. While a variety of definitions of aggression exist, in this paper we see it as any kind of behavior that has a disturbing effect on others. We have collected multi- and unimodal assessments by humans, who have given aggression scores on a 3 point scale. There are no trivial fusion algorithms to predict...
Detection and classification of significant human motions are important tasks when analyzing a video that records human activities. Among various human motions, we consider that repetitious motions are specially important since they are usually results of activities with clear intentions. In this paper, we propose and evaluate a method that detects video segments that contain repetitious motions,...
Recently, attributes have been introduced to help object classification. Multi-task learning is an effective methodology to achieve this goal, which shares low-level features between attribute and object classifiers. Yet such a method neglects the constraints that attributes impose on classes which may fail to constrain the semantic relationship between the attribute and object classifiers. In this...
In this paper, BP (back propagation) neural network is applied to represent and recognize human activities from observed sensor sequences. The proper features for activity recognition are selected. Then, the model of BP neural network for activity recognition is established. The results show that BP neural network can recognize human activity successfully and achieve good recognition accuracy.
The article shows possibilities of obtaining anthropometrical data through the image analysis -- image processing. In our case, the subject of the analysis is a specific human image focused on the unique contours of the body, carrying relevant physical information about a person. The elaborating of these curves brings us information needed for defining of a workplace parameters.
This paper applies the data mining process to predict hypertension from patient medical records with eight other diseases. A sample with the size of 9862 cases has been studied. The sample was extracted from a real world Healthcare Information System database containing 309383 medical records. We observed that the distribution of patient diseases in the medical database is imbalanced. Under-sampling...
A story is defined as "an actor(s) taking action(s) that culminates in a resolution(s)." In this paper, we investigate the utility of standard keyword based features, statistical features based on shallow-parsing (such as density of POS tags and named entities), and a new set of semantic features to develop a story classifier. This classifier is trained to identify a paragraph as a "story,"...
This paper studies a synthesis of password to be easily identified and hardly forgot. A new synthesis method is proposed to construct a good passwords that satisfy both requirements. Our method focuses on the change of frequency of combined words. Each of two words has a high term frequency but the combination is not quite common and then the frequency of the combined words can give strong impression...
This paper describes a reading quality scoring system based on large vocabulary continuous speech recognition (LVCSR). Our previous scoring system was based on forced alignment. A disadvantage of forced alignment based system is it can hardly catch huge kinds of reading miscues, while LVCSR based system avoids this disadvantage. The most challenge was that the LVCSR recognition rate was low on our...
Current commercial anti-malware products fail to guarantee a 100% detection and prevention of malware. This paper proposes an evaluation framework called ATE (Anti-malware Technique Evaluator) that can be used to evaluate commercial anti-malware products. ATE identifies the vulnerabilities in anti-malware products by providing a set of requirements that must be fulfilled by the anti-malware product...
Sentence fragment has a wide range of applications, such as short text mining, flow diagram search based on label similarity and so on. Existing methods aren't entirely appropriate for measuring similarity between sentence fragments since they were originally designed for complete sentences or long texts. So we pay more attention to proper nouns which carry important information in sentence fragments...
Textual Entailment (TE) is the task of recognizing entailment, paraphrase, and contradiction relations between a given text pair. The goal of textual entailment research is to develop a core inference component that can be applied to various domains, such as IR or NLP. Since the domain that a TE system applies to may be different from its source domain, it is crucial to develop proper datasets for...
Inter-subject correspondence is an important aspect of multi-subject fMRI studies. Recently, a new approach, called hyperalignment, has shown very promising results in fMRI functional alignment. Hyperalignment is based on Procrustean rotations and is connected, mathematically, to canonical correlation analysis. We review the core details of each approach, relate them through an SVD analysis, and indicate...
Facilitated by the latest advances of information technologies, online human computing resources provide researchers unprecedented opportunities to resolve a class of real-world problems that are challenging even to the computer algorithms, and yet manageable to human intelligence if working units are well organized. A problem in this category is image labeling, recognizing and categorizing targets...
Using brain-computer interfaces (BCIs) to improve human performance has become a state-of-the-art research topic. The concept of collaborative BCIs, which aimed to use multi-brain computing to enhance human performance, was proposed recently. To further study the feasibility of collaborative BCIs, here we propose to develop an online collaborative BCI to accelerate human response to visual target...
Current microarray technologies are able to assay thousands of samples over million of SNPs simultaneously. Computational approaches have been developed to analyse a huge amount of data from microarray chips to understand sophisticated human genomes. The data from microarray chips might contain errors due to bad samples or bad SNPs. In this paper, we propose a method to detect bad SNPs from the probe...
One critical issue in indoor human tracking is the design of map-aid algorithms that exploit indoor layout information. Most of current works adopt similar map-aid calibration techniques that eliminate invalid particles, which means particles propagating in inhumane manner. However, we find that these techniques have two serious problems in common, which we name acute sample impoverishment and observation...
Color signatures, histograms and bag of colors are basic and effective strategies for describing the color content of images, for retrieving images by their color appearance or providing color annotation. In some domains, colors assume a specific meaning for users and the color-based classification and retrieval should mirror the initial suggestions given by users in the training set. For instance...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.