scholarly journals Intelligent Conversational Agent for Enhancement of Online Communication in Universities: An Overview of Kenyatta University

2021 ◽  
Vol 4 (2) ◽  
pp. 85-90
Author(s):  
Isaac Kuria ◽  
Harrison Njoroge
Keyword(s):  
Online Communication ◽  
Training Data ◽  
Extensive Literature ◽  
Conversational Agents ◽  
Language Understanding ◽  
Matching Algorithm ◽  
University Websites ◽  
Language Pattern ◽  
Primary Means ◽  
Using Data

University websites and online portals are the primary means through which potential students and other stakeholders find important information about an institution. University websites are essential to these organizations’ marketing and communication efforts. In this paper, focus has been put on the need to complement these websites with the use of an AI Chatbot (UniBot) in order to serve more efficiently. This study aims at performing an extensive literature survey on intelligent conversational agents and the feasibility of applying them in enhancing online communication in universities. The study utilizes an iterative – incremental methodology to aid in design and development of UniBot, using AIML (Artificial Intelligent Markup Language) Pattern matching algorithm on the Pandorabot (AIAAS) platform, to generate high quality training data, with which, the agents Natural Language Understanding (NLU) model is trained. The study also provides for training and testing the agent using data which is acquired from Online Communication, University Website department at Kenyatta University.

scholarly journals Human Annotated Dialogues Dataset for Natural Conversational Agents

2020 ◽  
Vol 10 (3) ◽  
pp. 762
Author(s):  
Erinc Merdivan ◽  
Deepika Singh ◽  
Sten Hanke ◽  
Johannes Kropf ◽  
Andreas Holzinger ◽  
...  
Keyword(s):  
Natural Language ◽  
Detailed Analysis ◽  
Human Perception ◽  
Natural Language Understanding ◽  
Industrial Applications ◽  
Benchmark Dataset ◽  
Conversational Agents ◽  
High Quality ◽  
Language Understanding ◽  
Major Drawback

Conversational agents are gaining huge popularity in industrial applications such as digital assistants, chatbots, and particularly systems for natural language understanding (NLU). However, a major drawback is the unavailability of a common metric to evaluate the replies against human judgement for conversational agents. In this paper, we develop a benchmark dataset with human annotations and diverse replies that can be used to develop such metric for conversational agents. The paper introduces a high-quality human annotated movie dialogue dataset, HUMOD, that is developed from the Cornell movie dialogues dataset. This new dataset comprises 28,500 human responses from 9500 multi-turn dialogue history-reply pairs. Human responses include: (i) ratings of the dialogue reply in relevance to the dialogue history; and (ii) unique dialogue replies for each dialogue history from the users. Such unique dialogue replies enable researchers in evaluating their models against six unique human responses for each given history. Detailed analysis on how dialogues are structured and human perception on dialogue score in comparison with existing models are also presented.

scholarly journals Pre-Trained Transformer-Based Language Models for Sundanese

2021 ◽  
Author(s):  
Wilson Wongso ◽  
Henry Lucky ◽  
Derwin Suhartono
Keyword(s):  
Natural Language ◽  
Text Classification ◽  
Training Data ◽  
Language Models ◽  
Classification Task ◽  
Language Understanding ◽  
Training Corpus ◽  
Low Resource ◽  
Corpus Size ◽  
Fine Tune

Abstract The Sundanese language has over 32 million speakers worldwide, but the language has reaped little to no benefits from the recent advances in natural language understanding. Like other low-resource languages, the only alternative is to fine-tune existing multilingual models. In this paper, we pre-trained three monolingual Transformer-based language models on Sundanese data. When evaluated on a downstream text classification task, we found that most of our monolingual models outperformed larger multilingual models despite the smaller overall pre-training data. In the subsequent analyses, our models benefited strongly from the Sundanese pre-training corpus size and do not exhibit socially biased behavior. We released our models for other researchers and practitioners to use.

scholarly journals Using machine learning to predict extreme events in complex systems

2019 ◽  
Vol 117 (1) ◽  
pp. 52-59 ◽  
Author(s):  
Di Qi ◽  
Andrew J. Majda
Keyword(s):  
Dynamical Systems ◽  
Learning Strategies ◽  
Extreme Events ◽  
Water Waves ◽  
Extreme Values ◽  
Training Data ◽  
Partition Functions ◽  
Grand Challenge ◽  
Natural Systems ◽  
Using Data

Extreme events and the related anomalous statistics are ubiquitously observed in many natural systems, and the development of efficient methods to understand and accurately predict such representative features remains a grand challenge. Here, we investigate the skill of deep learning strategies in the prediction of extreme events in complex turbulent dynamical systems. Deep neural networks have been successfully applied to many imaging processing problems involving big data, and have recently shown potential for the study of dynamical systems. We propose to use a densely connected mixed-scale network model to capture the extreme events appearing in a truncated Korteweg–de Vries (tKdV) statistical framework, which creates anomalous skewed distributions consistent with recent laboratory experiments for shallow water waves across an abrupt depth change, where a remarkable statistical phase transition is generated by varying the inverse temperature parameter in the corresponding Gibbs invariant measures. The neural network is trained using data without knowing the explicit model dynamics, and the training data are only drawn from the near-Gaussian regime of the tKdV model solutions without the occurrence of large extreme values. A relative entropy loss function, together with empirical partition functions, is proposed for measuring the accuracy of the network output where the dominant structures in the turbulent field are emphasized. The optimized network is shown to gain uniformly high skill in accurately predicting the solutions in a wide variety of statistical regimes, including highly skewed extreme events. The technique is promising to be further applied to other complicated high-dimensional systems.

scholarly journals Application of Data Mining Techniques for Medical Data Classification: A Review

2018 ◽  
Vol 150 ◽  
pp. 06003 ◽  
Author(s):  
Saima Anwar Lashari ◽  
Rosziati Ibrahim ◽  
Norhalina Senan ◽  
N. S. A. M. Taujuddin
Keyword(s):  
Data Mining ◽  
Data Classification ◽  
Past Research ◽  
Medical Data ◽  
Extensive Literature ◽  
Literature Analysis ◽  
Data Mining Techniques ◽  
Medical Data Classification ◽  
Using Data

This paper investigates the existing practices and prospects of medical data classification based on data mining techniques. It highlights major advanced classification approaches used to enhance classification accuracy. Past research has provided literature on medical data classification using data mining techniques. From extensive literature analysis, it is found that data mining techniques are very effective for the task of classification. This paper analysed comparatively the current advancement in the classification of medical data. The findings of the study showed that the existing classification of medical data can be improved further. Nonetheless, there should be more research to ascertain and lessen the ambiguities for classification to gain better precision.

Mechanical System Modelling Using Physics-Based and Data-Based Subsystems

2020 ◽  
Author(s):  
Siamak Arbatani ◽  
József Kövecses
Keyword(s):  
Mechanical System ◽  
Interaction Force ◽  
Point Of View ◽  
Training Data ◽  
System Modelling ◽  
Core Element ◽  
Interaction Forces ◽  
Force Systems ◽  
Using Data ◽  
Novel Concept

Abstract Mechanical systems have been traditionally represented using parametric physics-based models. In this work, we introduce a novel concept, in this part of the mechanical system is represented using data-based subsystem models, and the overall mechanical system model is composed of these data-based and other, physics-based subsystems. A core element is the interfacing of the subsystems, which gives rise to interaction forces. The interfacing problem is formulated in a way that makes it possible to give a general representation to the interaction forces. We demonstrate that from the point of view of the physics-based subsystems the important element is that the data-based models can represent the interaction force systems properly. The data-based subsystems are developed using deep recurrent neural networks, and the training data is generated based on simulations using the fully parametric physics-based model of the system. Such training data could also be obtained through physical experimentation.

scholarly journals Modelling a Conversational Agent with Complex Emotional Intelligence

2020 ◽  
Vol 34 (10) ◽  
pp. 13710-13711
Author(s):  
Billal Belainine ◽  
Fatiha Sadat ◽  
Hakim Lounis
Keyword(s):  
Emotional Intelligence ◽  
Natural Language ◽  
Natural Language Understanding ◽  
Conversational Agent ◽  
Conversational Agents ◽  
Language Understanding ◽  
Leading Role ◽  
Complex Emotions

Chatbots or conversational agents have enjoyed great popularity in recent years. They surprisingly perform sensitive tasks in modern societies. However, despite the fact that they offer help, support, and fellowship, there is a task that is not yet mastered: dealing with complex emotions and simulating human sensations. This research aims to design an architecture for an emotional conversation agent for long-text conversations (multi-turns). This agent is intended to work in areas where the analysis of users feelings plays a leading role. This work refers to natural language understanding and response generation.

scholarly journals Short-Term Load Forecasting for a Single Household Based on Convolution Neural Networks Using Data Augmentation

Energies ◽  
2019 ◽  
Vol 12 (18) ◽  
pp. 3560 ◽  
Author(s):  
Acharya ◽  
Wi ◽  
Lee
Keyword(s):  
Neural Networks ◽  
Data Augmentation ◽  
Load Forecasting ◽  
Training Data ◽  
Convolution Neural Networks ◽  
Forecasting Method ◽  
High Volatility ◽  
Using Data ◽  
Advanced Metering ◽  
Residential Electric Demand

Advanced metering infrastructure (AMI) is spreading to households in some countries, and could be a source for forecasting the residential electric demand. However, load forecasting of a single household is still a fairly challenging topic because of the high volatility and uncertainty of the electric demand of households. Moreover, there is a limitation in the use of historical load data because of a change in house ownership, change in lifestyle, integration of new electric devices, and so on. The paper proposes a novel method to forecast the electricity loads of single residential households. The proposed forecasting method is based on convolution neural networks (CNNs) combined with a data-augmentation technique, which can artificially enlarge the training data. This method can address issues caused by a lack of historical data and improve the accuracy of residential load forecasting. Simulation results illustrate the validation and efficacy of the proposed method.

scholarly journals Deaf identities in a multicultural setting: The Ugandan context

2015 ◽  
Vol 4 (1) ◽  
Author(s):  
Anthony B. Mugeere ◽  
Peter Atekyereza ◽  
Edward K. Kirumira ◽  
Staffan Hojer
Keyword(s):  
Social Services ◽  
Ethnic Diversity ◽  
Minority Group ◽  
The United States ◽  
Deaf Community ◽  
Structured Interviews ◽  
Socially Constructed ◽  
Lower Case ◽  
Primary Means ◽  
Using Data

Often located far apart from each other, deaf and hearing impaired persons face a multiplicity of challenges that evolve around isolation, neglect and the deprivation of essential social services that affect their welfare and survival. Although it is evident that the number of persons born with or acquire hearing impairments in later stages of their lives is increasing in many developing countries, there is limited research on this population. The main objective of this article is to explore the identities and experiences of living as a person who is deaf in Uganda. Using data from semi-structured interviews with 42 deaf persons (aged 19–41) and three focus group discussions, the study findings show that beneath the more pragmatic identities documented in the United States and European discourses there is a matrix of ambiguous, often competing and manifold forms in Uganda that are not necessarily based on the deaf and deaf constructions. The results further show that the country’s cultural, religious and ethnic diversity is more of a restraint than an enabler to the aspirations of the deaf community. The study concludes that researchers and policy makers need to be cognisant of the unique issues underlying deaf epistemologies whilst implementing policy and programme initiatives that directly affect them. The upper case ‘D’ in the term deaf is a convention that has been used since the early 1970s to connote a ‘socially constructed visual culture’ or a linguistic, social and cultural minority group who use sign language as primary means of communication and identify with the deaf community, whereas the lower case ‘d’ in deaf refers to ‘the audio logical condition of hearing impairment’. However, in this article the lower case has been used consistently.

Sign in / Sign up

Export Citation Format

Share Document