2019.03 Paper Weekly 2
-
The proposed system has been designed for realistic scenarios considering different cases of occlusion, illumination changes, and different climatic conditions; a real scenario (the International Pittsburgh Airport parking lot) has been targeted with the condition that existing parking security cameras can be used, avoiding the deployment of new cameras or other sensors infrastructures. For design and validation, a new multicamera data set has been recorded. The system is based on existing object detectors (the results of two of them are shown) and different proposed postprocessing stages. The results clearly show that the proposed system works correctly in challenging scenarios including almost total occlusions, illumination changes, and different weather conditions.
-
Traffic flows in the network follow spatiotemporal patterns and this characteristic is used to suppress the impact of missing or erroneous data. The application of multilayer perceptrons and deep learning networks using autoencoders for the prediction task is evaluated. Prediction sensitivity to false data is estimated using traffic data from an urban traffic network.
-
In this paper, a novel deep-learning-based traffic flow prediction method is proposed, which considers the spatial and temporal correlations inherently. A stacked autoencoder model is used to learn generic traffic flow features, and it is trained in a greedy layerwise fashion. To the best of our knowledge, this is the first time that a deep architecture model is applied using autoencoders as building blocks to represent traffic flow features for prediction. Moreover, experiments demonstrate that the proposed method for traffic flow prediction has superior performance.
-
- Motivation: Academic social network
- Research Gap: Information overload to find a friend
- Proposed method: heterogeneous network; information gain; regularization-based optimization
-
In this paper, we define family profiles as the tags describing the demographic characteristics of a family, such as size, the presence of children, income, and socioeconomic status. It is a subset of the customer profile.
-
Given that current traffic fatalities and injuries are rare events compared to vehicle miles traveled, we show that fully autonomous vehicles would have to be driven hundreds of millions of miles and sometimes hundreds of billions of miles to demonstrate their reliability in terms of fatalities and injuries. Under even aggressive testing assumptions, existing fleets would take tens and sometimes hundreds of years to drive these miles—an impossible proposition if the aim is to demonstrate their performance prior to releasing them on the roads for consumer use. These findings demonstrate that developers of this technology and third-party testers cannot simply drive their way to safety. Instead, they will need to develop innovative methods of demonstrating safety and reliability. And yet, the possibility remains that it will not be possible to establish with certainty the safety of autonomous vehicles. Uncertainty will remain. Therefore, it is imperative that autonomous vehicle regulations are adaptive—designed from the outset to evolve with the technology so that society can better harness the benefits and manage the risks of these rapidly evolving and potentially transformative technologies.
-
The estimates reveal that the restriction policy in Beijing does not have significant influence on individuals’ decisions to drive, as compared with the policy’s influence on public transit. The rule-breaking behavior is constant and pervasive. We found that 47.8% of the regulated car owners didn’t follow the restriction rules, and drove “illegally” to their destination places. On average, car owners who traveled during peak hours and/or for work trips, and whose destinations were farther away from the city center or subway stations, were more likely to break the driving restriction rules. Therefore, Beijing is probably in need of more comprehensive and palatable policy instruments (e.g., a combination of congestion tolls, parking fees, fuel taxes, and high-speed transit facilities) to effectively alleviate traffic congestion and air pollution.
-
Distance is relevant even in online contexts that do not conform to the two-sided market structure. Compared to psychic distances, geographic distance is the strongest predictor of network ties. The predictive power of geographic distance is substitutable by other distance measures.
-
Online discussions can be tracked geographically over time. A distributed geo-aware streaming LDA (SLDA) model has been developed. Evaluation was performed during the 2016 American presidential primary elections.
-
Deep learning is a representation-learning method with multiple levels for discovering intrinsic features to better represent user preferences. We analyzed users’ check-in behavior in detail and developed a deep learning model to integrate geographical and social influences for POI recommendation tasks. We used a semi-restricted Boltzmann machine to model the geographical similarity and a conditional layer to model the social influence. Experiments with real-world LBSNs showed that our method performed better than other state-of-the-art methods. Theoretically, our study contributes to the effective usage of data science and analytics for social recommender system design. In practice, our results can be used to improve the quality of personalized POI recommendation services for websites and applications.
-
Advanced turbidity prediction for operational water supply planning
We develop a system for predicting turbidity peaking events at a water company by using operational, meteorological and hydrogeological factors. We explore correlations and variable significance and confirm, in most instances, that there is non-linearity in the data. We conclude that machine learning techniques can be used to successfully predict turbidity peaking events with AUC values over 0.80 at five of six sites.