Approximate Truth Discovery via Problem Scale Reduction

Abstract

Many real-world applications rely on multiple data sources to provide information on their interested items. Due to the noises and uncertainty in data, given a specific item, the information from different sources may conflict. To make reliable decisions based on these data, it is important to identify the trustworthy information by resolving these conflicts, i.e., the truth discovery problem. Current solutions to this problem detect the veracity of each value jointly with the reliability of each source for every data item. In this way, the efficiency of truth discovery is strictly confined by the problem scale, which in turn limits truth discovery algorithms from being applicable on a large scale. To address this issue, we propose an approximate truth discovery approach, which divides sources and values into groups according to a userspecified approximation criterion. The groups are then used for efficient inter-value influence computation to improve the accuracy. Our approach is applicable to most existing truth discovery algorithms. Experiments on real-world datasets show that our approach improves the efficiency compared to existing algorithms while achieving similar or even better accuracy. The scalability is further demonstrated by experiments on large synthetic datasets.

Truth Discovery via Exploiting Implications from Multi-Source Data

“…We observe imbalanced distributions of positive and negative claims in most of the investigated datasets, due to the long-tail characteristic of the real-world datasets [8,15]:…”

Truth finding by reliability estimation on inconsistent entities for heterogeneous data sets

“…Observation specification. Wang et al proposed an approximate truth discovery approach which divides sources and values into groups according to a user specified approximation criterion, and uses the groups for efficient inter-value influence computation to improve the accuracy [28]. Shi et al proposed a probabilistic graphical model incorporating silent rate, false spoken rate and true spoken rate three measures to simultaneously infer the truth as well as source quality without any priori training involving ground truth answers [36].…”

Securing the Internet of Things in the Age of Machine Learning and Software-Defined Networking

“…On the other hand, existing algorithms were not designed to deal with the unprecedented number of devices that are expected to populate the IoT network. Specifically, it has been shown that the above approaches do not wellscale with the number of devices [111], and cannot directly be applied to the IoT environment. Accordingly, the research community has recently devoted significant effort on the design and assessment of algorithms that enjoy both low computational complexity and scalability.…”

“…Truth discovery is generally related to trustworthiness, as trusted devices are expected to generate reliable and accurate information. However, due to the exorbitant number of interconnected IoT devices, scalable trust discovery processes are required as traditional solutions for other networking applications are expected to fail [111]. For example, [111] leverages trustworthiness and clustering algorithms to provide a trust discovery mechanism that also relies on problem scale reduction to achieve scalability.…”