Search for: All records

Effectively integrating knowledge into end-to-end task-oriented dialog systems remains a challenge. It typically requires incorporation of an external knowledge base (KB) and capture of the intrinsic semantics of the dialog history. Recent research shows promising results by using Sequence-to-Sequence models, Memory Networks, and even Graph Convolutional Networks. However, current state-of-the-art models are less effective at integrating dialog history and KB into task-oriented dialog systems in the following ways: 1. The KB representation is not fully context-aware. The dynamic interaction between the dialog history and KB is seldom explored. 2. Both the sequential and structural information in the dialog history can contribute to capturing the dialog semantics, but they are not studied concurrently. In this paper, we propose a novel Graph Memory Network (GMN) based Seq2Seq model, GraphMemDialog, to effectively learn the inherent structural information hidden in dialog history, and to model the dynamic interaction between dialog history and KBs. We adopt a modified graph attention network to learn the rich structural representation of the dialog history, whereas the context-aware representation of KB entities are learnt by our novel GMN. To fully exploit this dynamic interaction, we design a learnable memory controller coupled with external KB entity memories to recurrently incorporate dialog history context into KB entities through a multi-hop reasoning mechanism. Experiments on three public datasets show that our GraphMemDialog model achieves state-of-the-art performance and outperforms strong baselines by a large margin, especially on datatests with more complicated KB information. 

We present the results of a research study in which participants were subjected to social engineering attacks via telephone, telephone scams, in order to determine the features of scams which peopleare most susceptible to. The study has involved 186 university participants who were attacked with one of 27 different attack scripts which span different independent variables including the pretext used and the method of elicitation. In order to ensure informed consent, each participant was warned that they would receive a scam phone call within 3 months. One independent variable used is the time between the warning and launching the scam. In spite of this warning, a large fraction of participants were still deceived by the scam. A limitation to research in the study of telephone scams is the lack of a dataset of real phone scams for examination. Each phone call in our study was recorded and we present the dataset of these recordings, and their transcripts. To our knowledge, there is no similar publicly-available dataset or phone scams. We hope that our dataset will support future research in phone scams and their detection.