Time filter

Source Type

Dallas, TX, United States

Shen J.,SMU | Hua X.-S.,Microsoft | Sargin E.,Google
MM 2013 - Proceedings of the 2013 ACM Multimedia Conference | Year: 2013

Empowered by advances in information technology, such as social media network, digital library and mobile computing, there emerges an ever-increasing amounts of multimedia data. As the key technology to address the problem of information overload, multimedia recommendation system has been received a lot of attentions from both industry and academia. This course aims to 1) provide a series of detailed review of state-of-the-art in multimedia recommendation; 2) analyze key technical challenges in developing and evaluating next generation multimedia recommendation systems from different perspectives and 3) give some predictions about the road lies ahead of us. Copyright © 2013 ACM.

News Article
Site: techcrunch.com

Palaround began its life as one of what’s now many companies attempting to be the “Tinder for finding friends” — a list that now includes Tinder’s dating app rival, Bumble, in fact. But recently, the startup began rolling out a new product focused instead on bringing the Tinder swipe model to private organizations. A pivot from the earlier general purpose friend finder, the idea with this new platform is to offer a swiping app for closed networks — meaning alumni organizations, conferences, private clubs, festivals, universities and even businesses. Effectively a white-labeling of the Palaround platform, the service offers these organizations a way to connect their members with each other, while also taking advantage of a familiar swipe interface like the one popularized by Tinder. And like Tinder-esque dating apps, the platform offers a variety of standard features, including one-click registration via Facebook, one-on-one chats and an algorithm that helps connect users by taking into consideration factors important to the app at hand — like proximity, in the same work field or those with friends in common, among other things. However, as a DIY Tinder of sorts, Palaround also includes features designed for these unique networks, like a group chat option, and other tools designed to help get users chatting, like icebreakers and personality quizzes. With group chat, users have the option to add all their connections to a single chat room called a “society.” Meanwhile, after every 20 people who users are shown, the app stops to ask a few questions. This seemingly lightweight feature actually helps the app learn a lot about its users, in order to suggest better matches. For example, it might ask you how you work best — “alone,” or “in a pair,” or “within a small group,” etc. It’s easy for organizations to get started with Palaround, explains founder Joel Kliksberg. After setting up an account, they can import their email list and their users then automatically receive an invite link. “The whole thing takes less than three minutes,” he says. Users download the main Palaround app from the App Store, and sign into their network. Not only is the process meant to be simple to set up, it’s also designed to be affordable, Kliksberg adds. “We’ve talked to a lot of these events and conference organizers and they tell us that, for some reason, their people aren’t connecting or they’re not connecting efficiently, and that building a community is absolutely critical for them,” he says. However, building their own enterprise application for this purpose could cost $250,000 to $500,000. Plus, apps would have ongoing maintenance costs associated with them. Palaround’s alternative is an app-as-a-service. Palaround’s pricing is still in flux, but it’s well under what a standalone app would cost. Currently, its “starter” solution is $29 per month, while the “business” tier is $99 per month. (Pricing is based on a variety of factors, like number of users, GB of data transfer and storage and more.) In the future, other premium features may come into play — like support for push notifications, the ability to write your own questions or the ability to connect to third-party systems like MailChimp, Eventbrite or Campaign Monitor, for example. Though only a few weeks post-pivot, Palaround already has 50 companies signed up to beta test the new platform, including the University of Miami, Toyota, Lexus, Cordel Wine Institute, SMU Mustangs, Y Combinator, Bizhaus and others. Some gyms, buildings and even two weddings have also signed up to use the service. In most cases, the companies are starting with a small group. For instance, the YC network will just include the recent YC Fellows and the last three batches. A recent YC Fellow grad itself, L.A.-based Palaround has raised a six-figure seed round to get off the ground. The startup says its solution will be available for broader testing in a few weeks after a few kinks are worked out. In the meantime, interested customers can reach out via its website for more information.

When the economy is flourishing, will consumers be more eco-conscious in their purchases, or will they succumb to the temptation for a bigger car or appliance? We can now measure the impact of the macro-economy on such consumer choices to a surprisingly detailed degree, thanks to the research by Assistant Professor Anirban Mukherjee at the Singapore Management University (SMU) Lee Kong Chian School of Business. Professor Mukherjee's data-driven research into this question earned him the best paper in the marketing analytics track and the conference overall, at the annual Australian & New Zealand Marketing Academy Conference 2015 in Sydney. For his paper, "Does economic growth lead to consumers purchasing more energy efficient appliances?", Professor Mukherjee worked closely with Professor Andre Bonfrer, whose research focuses on understanding consumer behaviour and implications for marketing decisions. Before this research, it was well established that macro-economic changes alter consumer choices, over and above an individual household's financial outlook. But relatively little was known about how economic conditions affect the thousands of decisions driving behaviour within industries—the trade-offs people make when choosing between brands and features during their purchase. Seeing this vital gap in knowledge, Professor Mukherjee decided to study consumer choices on household appliances such as fridges, freezes and clothes dryers. Although these are everyday items in our lives, purchasing them involves complex consumer choices with many dimensions at play. "We cannot measure the environmental attitudes and motivation of so many consumers, so we use efficiency as a proxy for eco-friendly buying," he says. Efficiency can be measured, for example, through product information on the number of energy conservation stars a washing machine or fridge might have. In partnership with market research firm GfK Australia and New Zealand, Professor Mukherjee used a new data modelling paradigm called structural modelling to analyse purchases in each category of appliances over eight years. Studying the massive amount of purchases and attempting to isolate the dynamic drivers behind them was a mammoth challenge, but it gave Professor Mukherjee incredible insights into consumer behaviour. "For example, we can determine how much more someone would pay for an appliance with three energy stars instead of two energy stars," he says. The result of all this number crunching? Perhaps a realistic, if not practical, portrayal of our eco-consciousness—the data showed that during economic upswings, consumers cash in on convenience rather than energy efficiency. "Normal washing machines take about 30 minutes to wash clothes, whereas the energy-efficient machines take about an hour longer," notes Professor Mukherjee, explaining the trade-off consumers tend to make when times are good. The findings are in line with studies of the car industry, says Professor Mukherjee, who points to literature showing that in the automobile industry, consumers' desire for more motor power made those vehicles more popular than those which are energy-efficient. "It is quite astounding that the family cars today have more horse-power than the sports cars of the 1980s." The results were not all gloomy—for example, people do choose energy-efficient refrigerators for their purchase. However, what could be the reason behind most consumers' "less eco-friendly" purchases? Professor Mukherjee says it is partly to do with how we think about ourselves. "When we buy cars we also buy intangibles—status or lifestyle. One statistic says that just three percent of SUVs are used off-road, despite what you see in the advertising." Importantly, the robust collection and modelling of data might help marketers and policy makers understand the changes in consumer behaviour they are observing, suggests Professor Mukherjee. "Because there are so many other things going on, if market or political observers see preferences for eco-friendly appliances rise or fall, it might not simply come from a rise in eco-consciousness." Such precise data about the impact of economic forecasts within industries is important for planning advertising and pricing decisions, says Professor Mukherjee. This is especially important in markets where the manufacturing is not local, such as Australia, where a six-month lag could exist between ordering and arrival. "Marketers need to know whether changes in the economy will cause a move towards preferences for efficiency or convenience." Professor Mukherjee is looking to the future for the next phase of his research, as he seeks to study the impact of large technological advances, especially electric cars. "Car pollution seems central to mitigating global pollution, at least in large cities, and we have the same trade-off between 'doing good' and convenience," he says. In the face of data that people tend to upgrade and make life more convenient when the economy is doing well, Professor Mukherjee is surprisingly upbeat about the possibility of steering consumers away from their self-interest towards eco-consciousness. Research and good marketing can influence this conversation, he argues, citing the success of documentaries such as Al Gore's Inconvenient Truth in changing attitudes about climate change. "Marketing is based on change, maybe not of fundamental needs, as much as how we express ourselves. It does not have to be that large cars denote high status. If electric car manufacturers, such as Tesla, have the right appeal, those who buy for status might choose electric cars over others.

Seeking to address such urban challenges is Assistant Professor Pradeep Varakantham of the Singapore Management University (SMU) School of Information Systems, who uses a variety of methods from artificial intelligence, machine learning, operations research and behavioural economics to dynamically and continuously match supply with demand. "Since we have large data sets for these matching problems, they provide ideal settings for us to understand the potential impact of our research on real-world problems," he shares. Part of what draws Professor Varakantham to the field of continuous matching is the real potential to improve the quality of life in urban environments, from reducing waiting times for public transportation to improving public safety. At the same time, these problems are also theoretically and practically challenging, capturing his fascination as a researcher. To find ways to deploy emergency vehicles, security patrols and law enforcement personnel in more effective ways, Professor Varakantham engages with the people who keep Singapore safe around the clock. These include agencies such as the Singapore Civil Defence Force (SCDF, responsible for emergency response vehicles such as ambulances and fire trucks) and the Singapore Police Force and Coast Guard (patrol cars and boats). Once a problem is identified, Professor Varakantham constructs a model that best represents the available data, and uses it to obtain optimal strategies for matching supply with demand. These strategies are then tested out on a fresh set of real-world data (or on simulated data if this is not available). In their 2015 paper "Risk based Optimization for Improving Emergency Medical Systems" in Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, Professor Varakantham and his SMU colleagues Professor Lau Hoong Chuin and Sandhya Saisubramanian and used data from SCDF to develop a model and matching algorithms on how to best match ambulances to base stations so that they can respond quickly to emergencies. "We have shown in simulations that we can improve response times by at least two minutes," he says, adding that the team will be working with SCDF to apply their findings. Another intriguing research question involved the scheduling of traffic police patrols, which Professor Varakantham and colleagues studied in a 2014 paper, "STREETS: Game-Theoretic Traffic Patrolling with Exploration and Exploitation," in Proceedings of Innovative Applications in Artificial Intelligence (IAAI) at the Twenty-Eighth AAAI Conference on Artificial Intelligence. Traffic police patrols improve road safety by discouraging reckless driving. But with a limited number of personnel and a large number of road segments, the goal of the study was to reduce traffic violations in Singapore by continuously matching the two. "This is a fascinating problem because it can be viewed as a game between traffic police and drivers," explains Professor Varakantham. Drivers are more likely to disregard traffic laws if they can predict when traffic police would not be on a certain road segment. on the other hand, they are extra careful once traffic police are seen on the road. Using game theory concepts to model randomised strategies, the researchers developed a new algorithm they called STrategic Randomisation with Exploration and Exploitation in Traffic patrol Schedules, or STREETS. Their approach addresses a number of challenges for the first time, including the massive scale (thousands of drivers) and complexity (road networks are dynamic, and certain factors such as congestion and traffic signals are unpredictable) of the problem. "We proposed the use of randomised patrols that make it difficult for people to predict positions of traffic police at different times of the day, while also guaranteeing coverage of road segments that have high violations," he says. Professor Varakantham has used a similar approach to schedule randomised security patrols in rail networks; his approach has been tested in train stations in Los Angeles. In addition, a part of his collaborative work on reducing customer waiting times in theme parks has also been tested at a major theme park in Singapore. A major challenge for his field, says Professor Varakantham, is predicting demand. In some cases, there is not enough data to make strong predictions. For instance, it is difficult to find out how many violations occurred on a section of road that is unmonitored by traffic police. Prediction accuracy, Professor Varakantham believes, can be improved with a combination of machine learning techniques and additional sensors. Tapping into the network of communications between physical objects such as phones and cars, termed the Internet of Things, will help to reduce the human inefficiencies and address problems such as congestion, high response times and high vulnerability, he says. "In essence, the vision is to build a coordinated Internet of Things and People through the use of smart 'matching' algorithms that is extremely efficient." There is a balance to be struck, though, and some complex, high-stakes problems such as security will still require human experience. "Given the extreme nature of things that can happen if an error is made, is it possible for our systems to automatically figure out when to yield control to experienced humans?" he asks. "Can matching systems automatically self-analyse, explain and adjust their autonomy online? Undeniably, human expertise is still necessary, and what I'm doing is to make sure that the systems are performing at their best to support human experience." Explore further: Scientists investigate using artificial intelligence for next-generation traffic control More information: "Risk based Optimization for Improving Emergency Medical Systems" in Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, www.mysmu.edu/faculty/pradeepv/Papers/Ambulance.pdf

News Article | March 1, 2006
Site: www.techworld.com

NetApp has launched software to search backup and snapshot files for content as if they were Word documents. SnapSearch and Recover can search through hundreds-to-thousands of point-in-time copies and disk-to-disk backups online in NetApp SnapShot and SnapVault products. Users and sysadmins use a familiar web search interface to locate documents, presentations, etc. by subject, key word or other attributes and restore them in minutes. This enables sysadmins to meet compliance or legal discovery requests by treating thousands of consolidated online backups as a single search target. This is roughly similar to the way Google Desktop treats all of a user's files on their PC as a single search space. Users no longer need to know the precise location or specific name of a file in order to perform file recovery. Currently sysadmins would have to search each backup set and snapshot container individually. SnapSearch and Recover software can also be used to classify files into different types and migrate less-needed data to secondary storage. It is integrated into Kazeon's Information Server IS1200 product. NetApp is rebadging the IS1200 from Kazeon and it will cost $10,000 (about £6,000). Customers can purchase the software and NetApp hardware in a single system for about $60,000 (around £36,000). Kazeon has also linked its IS1200 product to Google enterprise search.

Discover hidden collaborations