News Article | December 5, 2016
Zonar, the leader in smart fleet management technology, announced today that Ian McKerlich has taken over as president and CEO. McKerlich, who had been serving as president at Zonar, succeeds Brett Brinton in the CEO role as part of a planned leadership transition resulting from the majority investment stake in Zonar by Continental AG. Brinton remains with the organization as a strategic advisor. “The entire team at Zonar is focused on continued innovation and modernization of connected intelligence solutions for commercial transportation so that we meet fleets’ evolving safety, efficiency and compliance needs,” said Ian McKerlich, president and CEO at Zonar. “With an expanding leadership team and support from Continental AG, we’ve dramatically improved our ability to do just that.” In addition to McKerlich’s new role, Zonar has promoted technology industry veteran Larry Jordan to senior vice president of product management. Jordan is responsible for managing all Zonar products across their lifecycle. One example is the Zonar Connect tablet and platform announced in October, which empowers commercial fleets to take greater control of their operations. Zonar has also made key technology hires to continue its path of delivering smart fleet technology solutions to its growing customer base. Arun Jacob has been hired as vice president of software development and Chad Maglaque as vice president of program and partner management. Jacob joins Zonar as a seasoned technology executive who has built and led the delivery of world-class products and services at companies including Hewlett-Packard, Walt Disney Company, Evri.com, Vulcan and others. At Zonar he will be in charge of software development, where he will lead engineering teams to deliver Zonar applications and APIs. Maglaque joins Zonar as an award-winning product and technology leader with over 20 years of experience managing large-scale products and services for both fast-paced start-ups as well as Fortune 100 companies, including Apple, Microsoft, Walmart Labs, Macy's and Internet-pioneer RealNetworks. At Zonar Maglaque will be responsible for program management, product delivery and managing the company’s partner programs. “Adding experienced leaders such as Chad and Arun will improve our ability to quickly deliver Zonar products to customers and work with partners and application developers to make sure we meet all of their fleet management technology needs,” McKerlich said. About Zonar Founded in 2001, Zonar has pioneered smart fleet management technology by providing innovative technology that has changed fleet operations in the vocational, pupil and commercial trucking industries. With a unique focus on this field, the Company offers a complete suite of solutions and specialized platforms for our customers in multiple markets. Our patented, award-winning technology keeps fleet owners and managers connected to their fleets and drivers to dispatchers. Headquartered in Seattle, Zonar also has a Technology Development Center in downtown Seattle, a regional office in Cincinnati, and a distribution center outside of Atlanta. For more information about Zonar Systems, go to http://www.zonarsystems.com
Li Y.,Walmart Labs |
Veeravalli V.V.,University of Illinois at Urbana - Champaign
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings | Year: 2016
Multiple observation sequences are collected, among which there is a small subset of outliers. A sequence is considered an outlier if the observations therein are generated by a mechanism different from that generating the observations in the majority of sequences. In the universal setting, the goal is to identify all the outliers without any knowledge about the underlying generating mechanisms. In prior work, this problem was studied as a universal hypothesis testing problem, and a generalized likelihood test was constructed and its asymptotic performance characterized. Here a connection is made between the generalized likelihood test and clustering algorithms from machine learning. It is shown that the generalized likelihood test is equivalent to combinatorial clustering over the probability simplex with the Kullback-Leibler divergence being the dissimilarity measure. Applied to synthetic data sets for outlier hypothesis testing, the performance of the generalized likelihood test is shown to be superior to that of a number of other clustering algorithms for sufficiently large sample sizes. © 2016 IEEE.
Duan H.,University of Illinois at Urbana - Champaign |
Zhai C.X.,University of Illinois at Urbana - Champaign |
Cheng J.,Walmart Labs |
Gattani A.,Walmart Labs
Proceedings of the VLDB Endowment | Year: 2013
The ability to let users search for products conveniently in product database is critical to the success of e-commerce. Although structured query languages (e.g. SQL) can be used to effectively access the product database, it is very dificult for end users to learn and use. In this paper, we study how to optimize search over structured product entities (represented by specifications) with keyword queries such as \cheap gaming laptop". One major dificulty in this problem is the vocabulary gap between the specifications of products in the database and the keywords people use in search queries. To solve the problem, we propose a novel probabilistic entity retrieval model based on query generation, where the entities would be ranked for a given keyword query based on the likelihood that a user who likes an entity would pose the query. Different ways to estimate the model parameters would lead to different variants of ranking functions. We start with simple estimates based on the specifications of entities, and then leverage user reviews and product search logs to improve the estimation. Multiple estimation algorithms are developed based on Maximum Likelihood and Maximum a Posteriori estimators. We evaluate the proposed product entity retrieval models on two newly created product search test collections. The results show that the proposed model significantly outperforms the existing retrieval models, benefiting from the modeling of attribute-level relevance. Despite the focus on product retrieval, the proposed modeling method is general and opens up many new opportunities in analyzing structured entity data with unstructured text data. We show the proposed probabilistic model can be easily adapted for many interesting applications including facet generation and review annotation. © 2013 VLDB Endowment.
Hindi H.,Walmart Labs
Proceedings of the American Control Conference | Year: 2013
Every year, thousands of cancer patients receive radiation treatment. Radiation beams are delivered to the patient from different directions, with high precision, with the objective of maximizing dosage to the tumor, while minimizing damage to the surrounding healthy tissue. This paper is an introduction to the basic optimization problem underlying radiation treatment planning. Specifically, we show how the computation of optimal beam directions and intensities can be formulated as a convex optimization problem. We discuss some common metrics and constraints used in radiation treatment planning, using methods from optimization, medical physics, and finance and risk management. We also review some effective parallelizable methods that have been developed for solving this inherently large-scale, multi-objective optimization problem. © 2013 AACC American Automatic Control Council.
Wu Y.,Simon Fraser University |
Liu X.,CAS Institute of Automation |
Xie M.,Walmart Labs |
Ester M.,Simon Fraser University |
Yang Q.,CAS Institute of Automation
WSDM 2016 - Proceedings of the 9th ACM International Conference on Web Search and Data Mining | Year: 2016
Collaborative Filtering (CF) is the most popular method for recommender systems. The principal idea of CF is that users might be interested in items that are favorited by similar users, and most of the existing CF methods measure users' preferences by their behaviors over all the items. However, users might have different interests over different topics, thus might share similar preferences with different groups of users over different sets of items. In this paper, we propose a novel and scalable method CCCF which improves the performance of CF methods via user-item co-clustering. CCCF first clusters users and items into several subgroups, where each subgroup includes a set of like-minded users and a set of items in which these users share their interests. Then, traditional CF methods can be easily applied to each subgroup, and the recommendation results from all the subgroups can be easily aggregated. Compared with previous works, CCCF has several advantages including scalability, flexibility, interpretability and extensibility. Experimental results on four real world data sets demonstrate that the proposed method significantly improves the performance of several state-of-the-art recommendation algorithms. © 2015 Copyright held by the owner/author(s).
Datta S.,Bell Laboratories |
Majumder A.,Bell Laboratories |
Naidu K.V.M.,Walmart Labs.
Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining | Year: 2012
In a team formation problem, one is required to find a group of users that can match the requirements of a collaborative task. Example of such collaborative tasks abound, ranging from software product development to various participatory sensing tasks in knowledge creation. Due to the nature of the task, team members are often required to work on a co-operative basis. Previous studies [1, 2] have indicated that co-operation becomes effective in presence of social connections. Therefore, effective team selection requires the team members to be socially close as well as a division of the task among team members so that no user is overloaded by the assignment. In this work, we investigate how such teams can be formed on a social network. Since our team formation problems are proven to be NP-hard, we design efficient approximate algorithms for finding near optimum teams with provable guarantees. As traditional data-sets from on-line social networks (e.g. Twitter, Facebook etc) typically do not contain instances of large scale collaboration, we have crawled millions of software repositories spanning a period of four years and hundreds of thousands of developers from GitHub, a popular open-source social coding network. We perform large scale experiments on this data-set to evaluate the accuracy and efficiency of our algorithms. Experimental results suggest that our algorithms achieve significant improvement in finding effective teams, as compared to naive strategies and scale well with the size of the data. Finally, we provide a validation of our techniques by comparing with existing software teams. © 2012 ACM.
Venkataswamy M.,Walmart Labs |
Agarwal N.,University of Arkansas at Little Rock |
19th Americas Conference on Information Systems, AMCIS 2013 - Hyperconnected World: Anything, Anywhere, Anytime | Year: 2013
User specific information in social media is sensitive and subject to privacy. Continuously changing privacy policies and configuration procedures in social media require users to constantly educate themselves of the changes. A collective intelligence driven approach, known as Collective-Context Based Privacy Model (C-CBPM) has been developed that recommends privacy policies based on community and trust gleaned from social network information. By defining userspecified contexts, C-CBPM advances the existing content, user, or role-based privacy models. This research examines the efficacy of C-CBPM using Facebook data comprising of 957,359 users, 957,357 connections, and 32,176 communities. Objective trust and privacy risk assessment measures are developed. Results indicate promising findings with 83% correct recommendations. Out of the 17% incorrect recommendations, almost all (i.e., 99.24% of the incorrect recommendations) incur only 25% risk and only 0.018% incur 100% or maximum risk, in the worst-case scenario. The results demonstrate the feasibility of the C-CBPM in real world for community driven privacy recommendations. © (2013) by the AIS/ICIS Administrative Office All rights reserved.
Chen J.,General Electric |
Tang L.,Walmart Labs |
Liu J.,Siemens AG |
Ye J.,Arizona State University
IEEE Transactions on Pattern Analysis and Machine Intelligence | Year: 2013
In this paper, we consider the problem of learning from multiple related tasks for improved generalization performance by extracting their shared structures. The alternating structure optimization (ASO) algorithm, which couples all tasks using a shared feature representation, has been successfully applied in various multitask learning problems. However, ASO is nonconvex and the alternating algorithm only finds a local solution. We first present an improved ASO formulation ((iASO)) for multitask learning based on a new regularizer. We then convert (iASO), a nonconvex formulation, into a relaxed convex one ((rASO)). Interestingly, our theoretical analysis reveals that (rASO) finds a globally optimal solution to its nonconvex counterpart (iASO) under certain conditions. (rASO) can be equivalently reformulated as a semidefinite program (SDP), which is, however, not scalable to large datasets. We propose to employ the block coordinate descent (BCD) method and the accelerated projected gradient (APG) algorithm separately to find the globally optimal solution to (rASO); we also develop efficient algorithms for solving the key subproblems involved in BCD and APG. The experiments on the Yahoo webpages datasets and the Drosophila gene expression pattern images datasets demonstrate the effectiveness and efficiency of the proposed algorithms and confirm our theoretical analysis. © 1979-2012 IEEE.
Choe T.E.,ObjectVideo |
Deng H.,ObjectVideo |
Guo F.,Walmart Labs |
Lee M.W.,Intelligent Automation Inc. |
Proceedings of the IEEE International Conference on Computer Vision | Year: 2013
We propose a novel video event retrieval algorithm given a video query containing grouped events from large scale video database. Rather than looking for similar scenes using visual features as conventional image retrieval algorithms do, we search for the similar semantic events (e.g. finding a video such that a person parks a vehicle and meets with other person and exchanges a bag). Videos are analyzed semantically and represented by a graphical structure. Now the problem is to match the graph with other graphs of events in the database. Since the query video may include noisy activities or some event may not be detected by the semantic video analyzer, exact graph matching does not always work. For efficient and effective solution, we introduce a novel sub graph indexing and matching scheme. Sub graphs are grouped and their importance is further learned over video by topic learning algorithms. After grouping and indexing sub graphs, the complex graph matching problem becomes simple vector comparison in reduced dimension. The performances are extensively evaluated and compared with each approach. © 2013 IEEE.
Cao N.,Walmart Labs |
Wang C.,City University of Hong Kong |
Li M.,Utah State University |
Ren K.,State University of New York at Buffalo |
Lou W.,Virginia Polytechnic Institute and State University
IEEE Transactions on Parallel and Distributed Systems | Year: 2014
With the advent of cloud computing, data owners are motivated to outsource their complex data management systems from local sites to the commercial public cloud for great flexibility and economic savings. But for protecting data privacy, sensitive data have to be encrypted before outsourcing, which obsoletes traditional data utilization based on plaintext keyword search. Thus, enabling an encrypted cloud data search service is of paramount importance. Considering the large number of data users and documents in the cloud, it is necessary to allow multiple keywords in the search request and return documents in the order of their relevance to these keywords. Related works on searchable encryption focus on single keyword search or Boolean keyword search, and rarely sort the search results. In this paper, for the first time, we define and solve the challenging problem of privacy-preserving multi-keyword ranked search over encrypted data in cloud computing (MRSE). We establish a set of strict privacy requirements for such a secure cloud data utilization system. Among various multi-keyword semantics, we choose the efficient similarity measure of "coordinate matching," i.e., as many matches as possible, to capture the relevance of data documents to the search query. We further use "inner product similarity" to quantitatively evaluate such similarity measure. We first propose a basic idea for the MRSE based on secure inner product computation, and then give two significantly improved MRSE schemes to achieve various stringent privacy requirements in two different threat models. To improve search experience of the data search service, we further extend these two schemes to support more search semantics. Thorough analysis investigating privacy and efficiency guarantees of proposed schemes is given. Experiments on the real-world data set further show proposed schemes indeed introduce low overhead on computation and communication. © 2014 IEEE.