In this work, a new approach to fully automatic color image segmentation, called JSEG, is presented. First, colors in the are quantized several representing classes that can be used differentiate regions image. Then, pixel replaced by their corresponding class labels, thus forming class-map of A criterion for "good" segmentation using proposed. Applying local windows results "J-image", which high and low values correspond possible region boundaries centers, respectively. growing method then...

Local differential privacy (LDP) is a recently proposed standard for collecting and analyzing data, which has been used, e.g., in the Chrome browser, iOS macOS. In LDP, each user perturbs her information locally, only sends randomized version to an aggregator who performs analyses, protects both users against private leaks. Although LDP attracted much research attention recent years, majority of existing work focuses on applying complex data and/or analysis tasks. this paper, we point out...

Recommender systems are collecting and analyzing user data to provide better experience. However, several privacy concerns have been raised when a recommender knows user's set of items or their ratings. A number solutions suggested improve legacy systems, but the existing in literature can protect either ratings only. In this paper, we propose system that protects both For this, develop novel matrix factorization algorithms under local differential (LDP). with LDP, individual users randomize...

A compact color descriptor and an efficient indexing method for this are presented. The target application is similarity retrieval in large image databases using color. Colors a given region clustered into small number of representative colors. feature consists the colors their percentages region. measure similar to quadratic histogram distance defined descriptor. can be indexed three-dimensional (3-D) space thus avoiding high-dimensional problems associated with traditional histogram. For...

Organizations with a large user base, such as Samsung and Google, can potentially benefit from collecting mining users' data. However, doing so raises privacy concerns, risks accidental breaches serious consequences. Local differential (LDP) techniques address this problem by only randomized answers each user, guarantees of plausible deniability; meanwhile, the aggregator still build accurate models predictors analyzing amounts So far, existing LDP solutions either have severely restricted...

A mobile operating system often needs to collect frequent new terms from users in order build and maintain a comprehensive dictionary. Collecting keyboard usage data, however, raises privacy concerns. Local differential (LDP) has been established as strong standard for collecting sensitive information users. Currently, the best known solution LDP-compliant term discovery transforms problem into n-grams under LDP, subsequently reconstructs collected by modelling latter graph, identifying...

Background The of upper extremity closed kinetic chain exercise combined with biofeedback requires evidence-based guidelines to elucidate its impact on the proprioception, muscle strength, and function stroke patients. Objective aim this study is compare effects Methods 24 patients were randomly divided into two groups: group (UCKCBG; n = 11) control (CG; 13). Training was conducted five times a week for four weeks. Outcome measures included Thumb Localization Test (TLT), Medical Research...

We investigate the theoretical properties of robust estimators for regression coefficient function in functional linear regression. Robust procedure is provided where we use outlier-resistant loss functions problem, including non-convex functions. These estimates are computed using an iteratively reweighted penalized least-squares algorithm. Using pseudo data approach, able to show that our also achieve same convergence rate both prediction and estimation as least squares estimator classical...

Keystroke data collected from smart devices includes various sensitive information about users. Collecting and analyzing such raise serious privacy concerns. Google Apple have recently applied local differential (LDP) to address issue on learning new words users' keystroke data. However, these solutions require multiple LDP reports for a single word, which result in inefficient use of budget high computational cost. In this paper, we develop novel algorithm under LDP. Unlike the existing...

This paper investigates the use of through-skull sound conduction to authenticate smartglass users. We mount a surface transducer on right mastoid process play cue signals and capture skull-transformed audio responses through contact microphones various skull locations. resultant bio-acoustic information as classification features. In an initial single-session study (N=25), we achieved mean Equal Error Rates (EERs) 5.68% 7.95% with brow left process. Combining two substantially improves...

This study examines the efficacy and moderators of New York State interventions for schools in need improvement under NCLB, including: (1) school transfer, (2) supplementary education service (3) corrective action, (4) planning restructuring, (5) restructuring. Despite fact that increasingly aggressive treatment groups had higher performance gains relative to good standing, propensity score matching analysis results reveal negative or null effects interventions. There are indications effect...

Local differential privacy (LDP) is a recently proposed standard for collecting and analyzing data, which has been used, e.g., in the Chrome browser, iOS macOS. In LDP, each user perturbs her information locally, only sends randomized version to an aggregator who performs analyses, protects both users against private leaks. Although LDP attracted much research attention recent years, majority of existing work focuses on applying complex data and/or analysis tasks. this paper, we point out...