A dynamic goal adapted task oriented dialogue agent

Purpose Existing virtual agents (VAs) present in dialogue systems are either information retrieval based or static goal-driven. However, in real-world situations, end-users might not have a known and fixed goal beforehand for the task, i.e., they may upgrade/downgrade/update their goal components in real-time to maximize their utility values. Existing VAs are unable to handle such dynamic goal-oriented situations. Methodology Due to the absence of any related dialogue dataset where such choice deviations are present, we have created a conversational dataset called Deviation adapted Virtual Agent(DevVA), with the manual annotation of its corresponding intents, slots, and sentiment labels. A Dynamic Goal Driven Dialogue Agent (DGDVA) has been developed by incorporating a Dynamic Goal Driven Module (GDM) on top of a deep reinforcement learning based dialogue manager. In the course of a conversation, the user sentiment provides grounded feedback about agent behavior, including goal serving action. User sentiment appears to be an appropriate indicator for goal discrepancy that guides the agent to complete the user’s desired task with gratification. The negative sentiment expressed by the user about an aspect of the provided choice is treated as a discrepancy that is being resolved by the GDM depending upon the observed discrepancy and current dialogue state. The goal update capability and the VA’s interactiveness trait enable end-users to accomplish their desired task satisfactorily. Findings The obtained experimental results illustrate that DGDVA can handle dynamic goals with maximum user satisfaction and a significantly higher success rate. The interaction drives the user to decide its final goal through the latent specification of possible choices and information retrieved and provided by the dialogue agent. Through the experimental results (qualitative and quantitative), we firmly conclude that the proposed sentiment-aware VA adapts users’ dynamic behavior for its goal setting with substantial efficacy in terms of primary objective i.e., task success rate (0.88). Practical implications In real world, it can be argued that many people do not have a predefined and fixed goal for tasks such as online shopping, movie booking & restaurant booking, etc. They tend to explore the available options first which are aligned with their minimum requirements and then decide one amongst them. The DGDVA provides maximum user satisfaction as it enables them to accomplish a dynamic goal that leads to additional utilities along with the essential ones. Originality To the best of our knowledge, this is the first effort towards the development of A Dynamic Goal Adapted Task-Oriented Dialogue Agent that can serve user goals dynamically until the user is satisfied.

Download Full-text

Convergence of Password Guessing to Optimal Success Rates

Entropy ◽

10.3390/e22040378 ◽

2020 ◽

Vol 22 (4) ◽

pp. 378 ◽

Cited By ~ 1

Author(s):

Hazel Murray ◽

David Malone

Keyword(s):

Success Rate ◽

Real World ◽

End Users ◽

Success Rates ◽

The Public

Password guessing is one of the most common methods an attacker will use for compromising end users. We often hear that passwords belonging to website users have been leaked and revealed to the public. These leaks compromise the users involved but also feed the wealth of knowledge attackers have about users’ passwords. The more informed attackers are about password creation, the better their password guessing becomes. In this paper, we demonstrate using proofs of convergence and real-world password data that the vulnerability of users increases as a result of password leaks. We show that a leak that reveals the passwords of just 1% of the users provides an attacker with enough information to potentially have a success rate of over 84% when trying to compromise other users of the same website. For researchers, it is often difficult to quantify the effectiveness of guessing strategies, particularly when guessing different datasets. We construct a model of password guessing that can be used to offer visual comparisons and formulate theorems corresponding to guessing success.

Download Full-text

Session-Level User Satisfaction Prediction for Customer Service Chatbot in E-Commerce (Student Abstract)

Proceedings of the AAAI Conference on Artificial Intelligence ◽

10.1609/aaai.v34i10.7259 ◽

2020 ◽

Vol 34 (10) ◽

pp. 13973-13974

Author(s):

Riheng Yao ◽

Shuangyong Song ◽

Qiudan Li ◽

Chao Wang ◽

Huan Chen ◽

...

Keyword(s):

Customer Service ◽

Real World ◽

User Satisfaction ◽

Experimental Results ◽

Practical Significance ◽

Proposed Model ◽

Questions And Answers ◽

The Relationship

This paper aims to predict user satisfaction for customer service chatbot in session level, which is of great practical significance yet rather untouched. It requires to explore the relationship between questions and answers across different rounds of interactions, and handle user bias. We propose an approach to model multi-round conversations within one session and take user information into account. Experimental results on a dataset from a real-world industrial customer service chatbot Alime demonstrate the good performance of our proposed model.

Download Full-text

Learning task-oriented grasping for tool manipulation from simulated self-supervision

The International Journal of Robotics Research ◽

10.1177/0278364919872545 ◽

2019 ◽

Vol 39 (2-3) ◽

pp. 202-216 ◽

Cited By ~ 4

Author(s):

Kuan Fang ◽

Yuke Zhu ◽

Animesh Garg ◽

Andrey Kurenkov ◽

Viraj Mehta ◽

...

Keyword(s):

Success Rate ◽

Real World ◽

Large Scale ◽

Learning Task ◽

Training Process ◽

Task Success ◽

Task Goals ◽

Task Oriented

Tool manipulation is vital for facilitating robots to complete challenging task goals. It requires reasoning about the desired effect of the task and, thus, properly grasping and manipulating the tool to achieve the task. Most work in robotics has focused on task-agnostic grasping, which optimizes for only grasp robustness without considering the subsequent manipulation tasks. In this article, we propose the Task-Oriented Grasping Network (TOG-Net) to jointly optimize both task-oriented grasping of a tool and the manipulation policy for that tool. The training process of the model is based on large-scale simulated self-supervision with procedurally generated tool objects. We perform both simulated and real-world experiments on two tool-based manipulation tasks: sweeping and hammering. Our model achieves overall 71.1% task success rate for sweeping and 80.0% task success rate for hammering.

Download Full-text

A Unified Dialogue Management Strategy for Multi-intent Dialogue Conversations in Multiple Languages

ACM Transactions on Asian and Low-Resource Language Information Processing ◽

10.1145/3461763 ◽

2021 ◽

Vol 20 (6) ◽

pp. 1-22

Author(s):

Tulika Saha ◽

Dhawal Gupta ◽

Sriparna Saha ◽

Pushpak Bhattacharyya

Keyword(s):

Reinforcement Learning ◽

Virtual Agents ◽

Dialogue System ◽

Dialogue Management ◽

Hierarchical Reinforcement Learning ◽

Learning Framework ◽

User Query ◽

Dialogue Manager ◽

Task Oriented ◽

Multiple Languages

Building Virtual Agents capable of carrying out complex queries of the user involving multiple intents of a domain is quite a challenge, because it demands that the agent manages several subtasks simultaneously. This article presents a universal Deep Reinforcement Learning framework that can synthesize dialogue managers capable of working in a task-oriented dialogue system encompassing various intents pertaining to a domain. The conversation between agent and user is broken down into hierarchies, to segregate subtasks pertinent to different intents. The concept of Hierarchical Reinforcement Learning, particularly options , is used to learn policies in different hierarchies that operates in distinct time steps to fulfill the user query successfully. The dialogue manager comprises top-level intent meta-policy to select among subtasks or options and a low-level controller policy to pick primitive actions to communicate with the user to complete the subtask provided to it by the top-level policy in varying intents of a domain. The proposed dialogue management module has been trained in a way such that it can be reused for any language for which it has been developed with little to no supervision. The developed system has been demonstrated for “Air Travel” and “Restaurant” domain in English and Hindi languages. Empirical results determine the robustness and efficacy of the learned dialogue policy as it outperforms several baselines and a state-of-the-art system.

Download Full-text

OFCOD: On the Fly Clustering Based Outlier Detection Framework

Data ◽

10.3390/data6010001 ◽

2020 ◽

Vol 6 (1) ◽

pp. 1

Author(s):

Ahmed Elmogy ◽

Hamada Rizk ◽

Amany M. Sarhan

Keyword(s):

Data Mining ◽

Image Processing ◽

Intrusion Detection ◽

Real Time ◽

Outlier Detection ◽

Real World ◽

Medical Data ◽

Experimental Results ◽

Real Time Applications ◽

Real World Datasets

In data mining, outlier detection is a major challenge as it has an important role in many applications such as medical data, image processing, fraud detection, intrusion detection, and so forth. An extensive variety of clustering based approaches have been developed to detect outliers. However they are by nature time consuming which restrict their utilization with real-time applications. Furthermore, outlier detection requests are handled one at a time, which means that each request is initiated individually with a particular set of parameters. In this paper, the first clustering based outlier detection framework, (On the Fly Clustering Based Outlier Detection (OFCOD)) is presented. OFCOD enables analysts to effectively find out outliers on time with request even within huge datasets. The proposed framework has been tested and evaluated using two real world datasets with different features and applications; one with 699 records, and another with five millions records. The experimental results show that the performance of the proposed framework outperforms other existing approaches while considering several evaluation metrics.

Download Full-text

Review Summary Generation in Online Systems: Frameworks for Supervised and Unsupervised Scenarios

ACM Transactions on the Web ◽

10.1145/3448015 ◽

2021 ◽

Vol 15 (3) ◽

pp. 1-33

Author(s):

Wenjun Jiang ◽

Jing Chen ◽

Xiaofei Ding ◽

Jie Wu ◽

Jiawei He ◽

...

Keyword(s):

Decision Making ◽

Real World ◽

Text Summarization ◽

Experimental Results ◽

Product Review ◽

Comprehensive Review ◽

Online Systems ◽

Real World Datasets ◽

Different Characteristics

In online systems, including e-commerce platforms, many users resort to the reviews or comments generated by previous consumers for decision making, while their time is limited to deal with many reviews. Therefore, a review summary, which contains all important features in user-generated reviews, is expected. In this article, we study “how to generate a comprehensive review summary from a large number of user-generated reviews.” This can be implemented by text summarization, which mainly has two types of extractive and abstractive approaches. Both of these approaches can deal with both supervised and unsupervised scenarios, but the former may generate redundant and incoherent summaries, while the latter can avoid redundancy but usually can only deal with short sequences. Moreover, both approaches may neglect the sentiment information. To address the above issues, we propose comprehensive Review Summary Generation frameworks to deal with the supervised and unsupervised scenarios. We design two different preprocess models of re-ranking and selecting to identify the important sentences while keeping users’ sentiment in the original reviews. These sentences can be further used to generate review summaries with text summarization methods. Experimental results in seven real-world datasets (Idebate, Rotten Tomatoes Amazon, Yelp, and three unlabelled product review datasets in Amazon) demonstrate that our work performs well in review summary generation. Moreover, the re-ranking and selecting models show different characteristics.

Download Full-text

A New No Equilibrium Fractional Order Chaotic System, Dynamical Investigation, Synchronization, and Its Digital Implementation

Inventions ◽

10.3390/inventions6030049 ◽

2021 ◽

Vol 6 (3) ◽

pp. 49

Author(s):

Zain-Aldeen S. A. Rahman ◽

Basil H. Jasim ◽

Yasir I. A. Al-Yasir ◽

Raed A. Abd-Alhameed ◽

Bilal Naji Alhasnawi

Keyword(s):

Adaptive Control ◽

Fractional Order ◽

Chaotic System ◽

Real World ◽

Experimental Results ◽

Chaotic Attractors ◽

State Variables ◽

Digital Implementation ◽

Dynamical Behaviors ◽

Real World Applications

In this paper, a new fractional order chaotic system without equilibrium is proposed, analytically and numerically investigated, and numerically and experimentally tested. The analytical and numerical investigations were used to describe the system’s dynamical behaviors including the system equilibria, the chaotic attractors, the bifurcation diagrams, and the Lyapunov exponents. Based on the obtained dynamical behaviors, the system can excite hidden chaotic attractors since it has no equilibrium. Then, a synchronization mechanism based on the adaptive control theory was developed between two identical new systems (master and slave). The adaptive control laws are derived based on synchronization error dynamics of the state variables for the master and slave. Consequently, the update laws of the slave parameters are obtained, where the slave parameters are assumed to be uncertain and are estimated corresponding to the master parameters by the synchronization process. Furthermore, Arduino Due boards were used to implement the proposed system in order to demonstrate its practicality in real-world applications. The simulation experimental results were obtained by MATLAB and the Arduino Due boards, respectively, with a good consistency between the simulation results and the experimental results, indicating that the new fractional order chaotic system is capable of being employed in real-world applications.

Download Full-text

Spatio-Temporal Modeling on Village Land Feature Change

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.204-208.2721 ◽

2012 ◽

Vol 204-208 ◽

pp. 2721-2725

Author(s):

Hua Ji Zhu ◽

Hua Rui Wu

Keyword(s):

Real World ◽

Temporal Change ◽

End Users ◽

Temporal Modeling ◽

The Real ◽

Change Type ◽

Feature Change ◽

Spatio Temporal ◽

The Village ◽

Land Changes

Village land continually changes in the real world. In order to keep the data up-to-date, data producers need update the data frequently. When the village land data are updated, the update information must be dispensed to the end-users to keep their client-databases current. In the real world, village land changes in many forms. Identifying the change type of village land (i.e. captures the semantics of change) and representing them in the data world can help end-users understand the change commonly and be convenient for end-users to integrate these change information into their databases. This work focuses on the model of the spatio-temporal change. A three-tuple model CAR for representing the spatio-temporal change is proposed based on the village land feature set before change and the village land feature set after change, change type and rules. In this model, the C denotes the change type. A denotes the attribute set; R denotes the judging rules of change type. The rule is described by the IF－THEN expressions. By the operations between R and A, the C is distinguished. This model overcomes the limitations of current methods. And more, the rules in this model can be easy realized in computer program.

Download Full-text

GazeEMD: Detecting Visual Intention in Gaze-Based Human-Robot Interaction

Robotics ◽

10.3390/robotics10020068 ◽

2021 ◽

Vol 10 (2) ◽

pp. 68

Author(s):

Lei Shi ◽

Cosmin Copot ◽

Steve Vanlanduit

Keyword(s):

Cognitive Load ◽

Real World ◽

Robotic Manipulator ◽

Similarity Score ◽

Human Robot Interaction ◽

Experimental Results ◽

The Other ◽

Robot Interaction ◽

Run Length ◽

The Real

In gaze-based Human-Robot Interaction (HRI), it is important to determine human visual intention for interacting with robots. One typical HRI interaction scenario is that a human selects an object by gaze and a robotic manipulator will pick up the object. In this work, we propose an approach, GazeEMD, that can be used to detect whether a human is looking at an object for HRI application. We use Earth Mover’s Distance (EMD) to measure the similarity between the hypothetical gazes at objects and the actual gazes. Then, the similarity score is used to determine if the human visual intention is on the object. We compare our approach with a fixation-based method and HitScan with a run length in the scenario of selecting daily objects by gaze. Our experimental results indicate that the GazeEMD approach has higher accuracy and is more robust to noises than the other approaches. Hence, the users can lessen cognitive load by using our approach in the real-world HRI scenario.

Download Full-text

Validation of a Musculoskeletal Digital Assessment Routing Tool (DART): Protocol for a Pilot Randomised Crossover Non-Inferiority Trial (Preprint)

10.2196/preprints.31541 ◽

2021 ◽

Author(s):

Cabella Lowe ◽

Harry Hanuman Sing ◽

William Marsh ◽

Dylan Morrissey

Keyword(s):

Clinical Trials ◽

Data Collection ◽

Real World ◽

User Satisfaction ◽

Allocation Concealment ◽

Trial Methodology ◽

Primary Analysis ◽

Digital Assessment ◽

System Usability Scale ◽

The Impact

BACKGROUND Musculoskeletal conditions account for 16% of global disability, resulting in a negative effect on millions of patients and an increasing burden on healthcare utilization. Digital technologies to improve health care outcomes and efficiency are considered a priority; however, innovations are rarely tested with sufficient rigor in clinical trials, the gold standard for clinical proof of safety and efficacy. We have developed a new musculoskeletal Digital Assessment Routing Tool (DART) that allows users to self-assess and be directed to the right care. DART requires validation in a real-world setting prior to implementation. OBJECTIVE This pilot study will assess the feasibility of a future trial by exploring key aspects of trial methodology, assess the procedures and collect exploratory data to inform the design of a definitive, randomized, crossover, non-inferiority trial to assess DART safety and effectiveness. METHODS We will collect data from 76 adult participants presenting to an NHS England GP practice with a musculoskeletal condition. Participants will complete both a DART assessment and a physiotherapist-led triage with the order determined by randomization. The primary analysis will involve an absolute agreement ICC (A,1) estimate with 95% confidence intervals between DART and the clinician for assessment outcomes sign-posting to condition management pathways. Data will be collected to allow analysis of participant recruitment and retention, randomization, allocation concealment, blinding, data collection process and bias. In addition, the impact of trial burden and potential barriers to intervention delivery will be considered. DART user satisfaction will be measured using the System Usability Scale. RESULTS A UK NHS ethics submission will be submitted during June 2021 and pending approval, recruitment will commence during August 2021 with data collection anticipated to last for 3 months. Results will be reported in a follow-up paper later in 2021. CONCLUSIONS This study will inform the design of a randomized controlled crossover non-inferiority study that will provide evidence concerning mHealth DART system clinical sign posting in an NHS setting prior to real-world implementation. Success should produce evidence of a safe, effective system with excellent usability, facilitating quicker and easier patient access to appropriate care while reducing the burden on primary and secondary care musculoskeletal services. This rigorous approach to mHealth system testing could be used as a guide for other developers of similar applications. CLINICALTRIAL This trial is registered with Clinical Trials number NCT04904029

Download Full-text