Recently emerged intelligent assistants on smartphones and home electronics (e.g., Siri and Alexa) can be seen as novel hybrids of domain-specific task- oriented spoken dialogue systems and open-domain non-task-oriented ones. To realize such hybrid dialogue systems, this paper investigates determining whether or not a user is going to have a chat with the system. To address the lack of bench- mark datasets for this task, we construct a new dataset consisting of 15, 160 utter- ances collected from the real log data of a commercial intelligent assistant (and will release the dataset to facilitate future re- search activity). In addition, we investi- gate using tweets and Web search queries for handling open-domain user utterances, which characterize the task of chat de- tection. Experiments demonstrated that, while simple supervised methods are ef- fective, the use of the tweets and search queries further improves the F 1 -score from
12 Read more
A knowledge-based expert system is suggested as the intelligent assistant of specialist for diagnosis of intestinal tissue atrophy. The facts of system are prepared as the input of inferential engine based on the expert knowledge, tests, physical examination and processing the images of damaged tissue. Processing the images of damaged tissue provides a new method based on processing the endoscopic images of atrophy tissues in small intestine through the new morphological structures. These masks move on the images with almost 85% of similarity and adaptation, and then track and estimate the density of designed objects in masks on the image. The
requirement and it is really a time consuming process to search for information through these traditional methods. A study of user behaviour reveals that the users usually expect results bang on point and are too lazy to navigate through different web pages looking for potential information. Also, it reveals that 90% of the users are unwilling to specify their intentions clearly or give a feedback about a web page (thumbs up or thumbs down) due to the time constraint. All these necessitates for an alternative and easy way of information retrieval based on individual’s interest. Hence, in order to solve the challenge of retrieving personalized information intelligent agents are being developed. An Intelligent agent is a program that does the tough job of information gathering automatically without any human intervention. The past decade has seen a tremendous tilt towards agent oriented technology. The success of such agents can be a breakthrough in the technological world and renders great scope ahead. In particular, the iAGENT is a program running on a client terminal that assists the user to get the relevant information. It acts like a stand-alone proxy, exactly like the WebMate , that monitors all the HTTP transactions taking places between the end user and the WWW. To give a brief overview of the entire approach, firstly in this paper, a brief study of user behaviour is made to find out the key points to be considered. Then we describe the procedure to build an initial user profile, update the naive profile and propose an algorithm for query modification in an entirely new way.
Here, we present the concept of an Intelligent Assis- tant Carer (IAC), a system that is aimed to address both issues. The system is designed to be modular, allowing the users or their carers to activate individual modules when required. Younger elderly users who live independently are already used to some ICT technologies; therefore, it is easier for them to start using the system. Initially, they might use it for leisure activities, such as an exer- cise tracker or as a means to stay connected with friends. Through time, the system “grows” with the users, mean- ing that the system learns about their habits and needs, and the users adopt the system as a part of their daily life. Therefore, the introduction of additional modules is a
10 Read more
S voice is a virtual mobile intelligent personal assistant used only in Samsung Smart phones. It is a knowledge navigator. It can run large number of tasks through voice commands. The advantage is that it saves time and effort to perform those functions from typing. This voice assistant is preloaded in the system. The application uses mainly natural language interface to answer questions and make recommendation and perform actions by delegating request to set of web services.
The safety control situation mainly refers to the comprehensive situation of missile's current position, trajectory, safety pipeline, protection target distribution, marine vessel distribution, etc. The purpose of the intelligent assistant decision system is to provide better assistant support for the commander, but it can't replace the commander to make decision. So the main function of the system is still to display the security situation in real time. On the one hand, the situation displaying needs to show all data comprehensively for the commander; on the other hand, it should push the results of auxiliary decision-making to the commander without interfering with the commander.
All these virtual assistants are designed with an aim to replace a human being and assist the people around. These systems are doing really well in their domain. In the world of technology, people are more comfortable in communicating with a computer or any other electronic device than a human being. Sometimes people hesitate to interact and ask queries to a human being as either they want to keep it secret or they feel shy. Sometimes they want somebody’s help but are not able to reach a right person to seek a help. This motivated us to design an interactive and intelligent student assistant situated in a close environment. It helps students who are new to the college environment. When a newcomer or a fresher enters in a college, he/she may not be familiar with the environment and seeks for the guidance from some experienced person like senior or teacher. But at the same time, some of them fear of being ragged or hesitate to ask. Hence, the new-comers remain unaware of the college environment. In such a scenario, the proposed student assistant, StuA, helps the newcomers and tries to provide the required information in a safe and sound manner to the students. It helps in familiarizing them with the rules and regulations of the college in a comfortable manner. StuA is equipped with the unique features which allow the user to ask questions on WHAT IS, WHERE, WHEN, WHAT HAPPEN format. Moreover, it facilitates real-time, low-cost expert-level assistance 24X7, unlike a human. The system based on its previously gained knowledge and beliefs is able to provide answers to most of the queries. The set of questions is not limited to only the prefixed questions. Nevertheless, to the best of our knowledge, no such virtual intelligent assistant exists till date. Moreover, it can be further customized to handle the queries of a new person in any environment such as a new office, organization, reception of hotels, hospitals, schools and malls.
In the near future, smart cooking assistants will be essential to everyone with the ever-improving intelligent systems. Thus, the proposed system will effectively and efficiently bridge the gap that is created due to the tedious task of collecting knowledge and information from various sources about various recipes. This system will only be taking the mere input of voice from the user asking for various recipes, and the system cooking assistant will be doing all the tasks of searching for the recipe that the user has asked for and providing the required content. The system will also the problem of language barrier
Abstract:- Modern Trip Assistant System is an advanced tour planning guide for tourists who wish to travel historical and popular tourist places in Sri Lanka. The system is designed to help tourists to plan a day-based trip with all the facilities. More specifically, the flow of the proposed system is when a user opens this application for the first time, a login/signup interface will be displayed. After that user has to specify the no of days or total duration of the trip. Then in the next screen user can choose all the tourist places and our system will plan the trip according to the tourist places. Our system will notify tourists according to the trip plan in order to prevent time lack. Duration for each tourist place will be determined with the help of the data provided by the tourist guides when registering to our system. At the end of the trip, the system will charge for the entire trip with all the expenses as a single payment. All the payment-related information can be accessed using the web interface of Modern Trip Assistant System.
The MADS data set (Multi-Applications Dia- logues for Smart phones) was constructed and used to develop the SLU modules for the intelli- gent personal assistant software. The MADS data set was annotated by 8 NEs, 28 targets, 5 opera- tors and 6 speech-act tags. In addition, The MADS data set consists of 1,925 user utterance in 6 domains: weather, clock, alarm, schedule, exchange and traffic. The Mallet toolkit was chosen for our CRF model (McCallum. 2002).
Master Thesis: Drum Assistant Page 38 be demonstrated by some virtual hands (4), because this was not feasible to implement. However, it is possible to play along with the exercise. The feedback methods that are implemented do not immediately show the errors made by the student (5). A future research could however implement and test a feedback method that is able to do this. A teacher cannot create new exercises within the application (6), but only specify in ABC- notation using a simple text editor. This is of course not very user-friendly, but the realization of a special editor was too ambitious for this project. A teacher can record their playing of the exercises to serve as a reference for analysis of the student’s performance (7). The student has the option to play along with these recordings. This kind of feedback is implemented in the application, but not tested in an experiment. It was not feasible to create the option to communicate with a teacher within the application (8). The performance scores of the student are stored in a file, but these cannot be viewed within the application (9). Because we kept the functionally limited, we also not implemented an option for a teacher to manage the accessible exercises (10). All in all, we can conclude that the functional requirements were a bit too ambitious to implement in our project. However, it certainly contains some useful additions to the system for future work.
51 Read more
Based on the current system on healthcare environment, the patient need to go to the hospital or clinic to know their health status and need to see the doctor if their health is in critical condition. There is no appropriated system for the patient to know their health condition while they are at home or any place they go. These are the problem that have been identified which lead to the development of Electronic Medical Assistant on Smartphone.
24 Read more
This main objective of this project is to study the technique concept, performance, and process a Library Assistant robot. So, it must be design a robot which has the capability to take the book and send it at their place. Besides that, it can be assisting handling to send the book at high place. This project is to design a robot has CCTV or camera for watching and has mechanism to transfer the books to it shelf for transfer the book by Personal Computer’s (PC) handling.
24 Read more
At beginning, our system will mimic like a real doctor, by asking which body part is being affected. The patient will need to select the body part from the displayed list. The second step as a doctor asks about is symptoms, similarly our personal health assistant will display a set of symptoms pertaining to those selected body parts. The patient would be able to select multiple symptoms that he or she is suffering with. On the basis of symptoms given by patient, the system will make predictions of the probable diseases in the form percentages. The predictions are made by our random forest classifier object which is trained beforehand on dataset of symptoms and disease . The patient would be able to see a list of predicted diseases given by our random forest classifier. Since the dataset present online is raw data, it must be pre-processed using various data pre-processing technique. Our dataset consisted of three columns- symptoms, diseases and weight. Here weight parameter signifies the number of instances of such diseases. The dataset is processed such that all the symptoms are turned into columns and last column would be of disease column. The symptoms pertaining to particular diseases are marked as 1, while rest 0. This processed data is now trained using random forest classification algorithm. The reason behind choosing random forest classifier, as machine learning algorithm, is its ability construct multiple decision trees on features and glue them together to get a more accurate and stable prediction. The forest that it builds is an ensemble of many decision trees which are trained using bagging method.
To ensure error-free operation, timing constraints have to be satisfied for the switching pattern that causes the worst- case delay, which are the ↑↓ and ↓↑ combinations. Since the effective load is highest for these patterns, the size of the buffer designed statically for the worst-case delay is much larger than would be necessary for the same timing requirements for other patterns . Now this worst-case condition occurs only twice out of 16 possible input switching patterns, with a probability of 1/8 for simultaneously switching lines if the transitions are equally distributed as in a random bit stream. For the 14 other cases, the wire is driven faster, which just translates to slack which typically cannot be used. The driver proposed here changes its drive strength depending on the neighbour’s switching direction by using some simple logic. A basic schematic of the proposed SMART repeater is shown in Figure 2. If the switching pattern belongs to Groups 1, 3, or 4, a single inverter (the Main driver) drives the interconnect. When a switching pattern in Group 5 occurs, another inverter (the Assistant) also drives the line, increasing the total drive strength appropriately.
Given the overall importance of transportation in urban development, many smart city technologies are focusing on developing intelligent transportation systems (ITS) to support the city growth. For example, some of the ITS strategies address the problem of reducing air pollution by promoting bike rental/sharing systems , , electric vehicles , autonomous cars  and taxi/buses on demand . Other strategies focus on developing efficient travel information systems , ticketing and mobile payments  to encourage people to use public transportation. Last, but not least, considerable effort is put into developing smart strategies for traffic planning. Here, traffic management systems , as well as smart parking applications  are developed with the purpose of monitoring, decongesting and directing drivers toward free parking places. All these strategies are fundamental for cities to become more sustainable, clean and safe living places.
The DSM has seen considerable use in the planning and management of design projects, however it has rarely been applied as an assistant to the enactment of a design project. This work indicates that the DSM may be used to provide considerable help to the designer in both understanding the design problem, and managing and enacting the design activity with the result of a considerable reduction in the time taken to generate suitable solutions especially in situations where the designer may not have a great understanding of either the Designer software of the associated model. These results are further improved with the use of a Genetic Algorithm to optimise the sequence of characteristics within the matrix with the aim of reducing the amount of iteration within the process, and reducing the time taken to satisfy the design requirements.
10 Read more
The modern world meets an innovation every day! Innovation is aimed at reducing human effort as well as providing a secure environment. With this in mind, we propose a Zigbee Network System which is centered on an Intelligent Robot that provides both security and control of the entire house. The Robot acts as a virtual assistant by monitoring the house to detect the presence of intruders and simultaneously providing electronic control of the entire house from a single place. The house is monitored through a wireless video camera mounted on the robot, which is in turn controlled through Zigbee IEEE 802.15.4 standard. The appliances are interfaced to the Robot using relays while the entire house is monitored and controlled through the Zigbee Network. Thus, the proposed robot serves as a simple, yet powerful, low cost alternative to the assistant robots by simultaneously providing Home Security and Home Automation.
Muhammad Sheikh Sadi, Saifuddin Mahmud, Md. Mostafa Kamal, Abu Ibne have proposed a solution for blind individuals to navigate securely by identifying the barrier and producing the associated warning signal as per the range of the barrier. The approach is given by the development of a moving assistant integrated in a spectacle glass between a barrier detection unit and also an alarm generator. There will only be one ultrasonic sensor in the barrier detection unit that can reach a range of 3 meters as well as an angle of 60 degrees to spot barriers. The barrier detection unit produces a large-frequency signal across an ultrasonic sensor and evaluates the signal return back by the sensor. Therefore, the length of the barrier is evaluated and so this data about the barrier is transmitted to the blind through the use of an alert generator that produces an alarm equivalent to the range.
In this era of digitization and automation, Internet of things (IoT) provides a platform that allows devices to be connected, sensed and controlled remotely across a network infrastructure. As an organization grows every year, new management problems and energy issues appear. Monitoring and controlling the unused devices that consume power during human absence is a major inability. In addition to this, coordinating the people participating in the daily activities of the laboratory is tedious when population of the usage of space out numbers a manageable threshold. This work targets to develop a smart laboratory system based on IoT and mobile application technologies to monitor the overall activities of the lab including energy consumption and utilization of devices, thereby providing a smart environment to the campus with enhanced security, energy efficiency and comfort. The aim is to control and monitor the things such as lights, fans, projector and air-conditioner of IoT lab in CIT campus using Google assistant or chat bot. The instigation of Google assistant not only provides a user-friendly access to the things but also makes the process interactive with its response. The use of voice commands overcomes the mundane task of switching the appliances. The status and energy consumption of the devices were viewed on a dashboard which is available in the website (www.citeceiot.in/dashboard/graph.php). From the results of implementation, it is observed that the appliances in the lab were remotely monitored and controlled thereby reducing power consumed and human energy considerably.