With the increasing availability of GPS-equipped mobile devices, location-based services have become an integral part of people's everyday life. Among one of the initial steps of positioning data management, map matching aims to reduce the uncertainty in a trajectory by matching the GPS points to the road network on a digital map. Most existing work has focused on estimating the likelihood of a candidate path based on the GPS observations, while neglecting to model the probability of a route choice from the perspective of drivers. Here we propose a novel feature-based map matching algorithm that estimates the cost of a candidate path based on both GPS observations and human factors. To take human factors into consideration is very important especially when dealling with low sampling rate data where most of the movement details are lost. Additionally, we simultaneously analyze a sebsequence of coherent GPS points by utilizing a new segment-based probabilistic map matching strategy, which is less susceptible to the noiseness of the positioning data. We have evaluated both the offline and the online versions of our proposed approach on a public large-scale GPS dataset, which consists of 100 trajectories distributed all over the world. The experimental results show that our method is robust to sparse data with large sampling intervals (e.g., 60 s - 300 s) and challenging track features (e.g., u-turns and loops). Our method obtains the state-of-the-art map matching accuracy with a competitive processing time compared with existing map matching approaches.
In marketing, helping manufacturers to find the matching preferences of potential customers for their products is an essential work especially in e-commerce analyzing with big data. The aggregate reverse rank query has been proposed to return top-$k$ customers who regard a given product bundling as highest aggregate rank than other customers, where the aggregate rank is defined as the sum of each product's rank. This query correctly reflects the request only when the customers consider the products in the product bundling equally. Unfortunately, rather than thinking products equally, in most cases, people buy a product bundling because they appreciate a special part of the bundling. Manufacturers, such as video games companies and cable television industries, are also willing to bundle some attractive products with less popular products for the purpose of maximum benefits or inventory liquidation. Inspired by the necessity of general aggregate reverse rank query for unequal thinking, we propose a weighted aggregate reverse rank query which treats the elements in product bundling with different weights to target customers from all aspects of thought. To solve this query efficiently, we first try a straightforward extension. Then we re-build the bound-and-filter framework for the weighted aggregate reverse rank query. We prove theoretically that the new approach finds the optimal bounds and develops the maximum efficient algorithm based on this bounds. The theoretical analysis and experimental results demonstrated the efficacy of the proposed methods.