No description available.

With the rapid development of industrialization and urbanization, industrial development and population growth drive the expansion of urban space, urban transportation demand shows the characteristics of spatial decentralization and diversification, and transportation travelers' requirements for mobility, accessibility, and comfort of transportation travel services are enhanced. Mobility on demand (MoD) services such as DiDi and Uber are new modes of public transportation, bringing many new opportunities and challenges. MoD travel services, shared bicycles, and other complementary public transport modes are rapidly developing in the "Internet +" environment, serving the "one mile" before and after the residents' travel. MoD technologies play an important role as a feeder to the main public transportation lines, helping to increase public transportation patronage and improve the speed of travel for residents. In this context, the study aims to develop a multi-modal public transportation system network design methodology to provide better operational coordination between different modes of transportation and to provide faster travel services. In order to promote better coordination between different transportation modes and to provide theoretical and methodological support for the development of a multi-modal public transportation system network design system, a bi-level planning model for this problem is first constructed. The upper-level planning model is used to minimize the total travel time and cost of passengers and the economic cost of public transportation operators, and to decide which bus lines to operate, the structure of bus lines, and the frequency of operating bus lines; the lower-level operating model is used to assign passengers to make travel mode choices and to carry out traffic distribution of the public transportation network based on the minimum number of interchanges. Then, based on this bi-level planning model, an improved genetic algorithm is developed to solve the upper-level public transportation network planning problem, in which the algorithm for passenger flow allocation in the lower-level planning model is nested in the genetic algorithm.  Finally, the developed methodology is validated for the benchmark Mandl network design by comparing with the traditional public transportation network. The results show that the multi-modal public transportation network can effectively reduce passenger travel time compared with the traditional public transportation network at similar costs. Finally, we applied the network design method for the Barkarby area in the north of Stockholm, Sweden. The results show that it is appropriate to allocate mobility on demand vehicles in this area. The constructed model and the proposed algorithm are scientifically valid and can provide theoretical methodological reference and decision support for engineering practice.

No description available.

No description available.

No description available.

No description available.

No description available.

The original VTQuest web-based software system requires users to interact using a mouse or a keyboard, forcing the users' hands and eyes to be constantly in use while communicating with the system. This prevents the user from being able to perform other tasks which require the user's hands or eyes at the same time. This restriction on the user's ability to multitask while using VTQuest is unnecessary and has been eliminated with the creation of the VTQuest Voice web-based software system. VTQuest Voice extends the original VTQuest functionality by providing the user with a voice interface to interact with the system using the Speech Application Language Tags (SALT) technology. The voice interface provides the user with the ability to navigate through the site, submit queries, browse query results, and receive helpful hints to better utilize the voice system. Individuals with a handicap that prevents them from using their arms or hands, users who are not familiar with the mouse and keyboard style of communication, and those who have their hands preoccupied need alternative communication interfaces which do not require the use of their hands. All of these users require and benefit from a voice interface being added onto VTQuest. Through the use of the voice interface, all of the system's features can be accessed exclusively with voice and without the use of a user's hands. Using a voice interface also frees the user's eyes from being used during the process of selecting an option or link on a page, which allows the user to look at the system less frequently. VTQuest Voice is implemented and tested for operation on computers running Microsoft Windows using Microsoft Internet Explorer with the correct SALT and Adobe Scalable Vector Graphics (SVG) Viewer plug-ins installed. VTQuest Voice offers a variety of features including an extensive grammar and out-of-turn interaction, which are flexible for future growth. The grammar offers ways in which users may begin or end a query to better accommodate the variety of ways users may phrase their queries. To accommodate for abbreviations of building names and alternate pronunciations of building names, the grammar also includes nicknames for the buildings. The out-of-turn interaction combines multiple steps into one spoken sentence thereby shortening the interaction and also making the process more natural for the user. The addition of a voice interface is recommended for web applications which a user may need to use his or her eyes and hands to multitask. Additional functionality which can be added later to VTQuest Voice is touch screen support and accessibility from cell phones, Personal Digital Assistants (PDAs), and other mobile devices. / Master of Science

Yes / Aggressive posts containing symbolic and offensive images, inappropriate gestures along with provocative textual comments are growing exponentially in social media with the availability of inexpensive data services. These posts have numerous negative impacts on the reader and need an immediate technical solution to filter out aggressive comments. This paper presents a model based on a Convolutional Neural Network (CNN) and Binary Particle Swarm Optimization (BPSO) to classify the social media posts containing images with associated textual comments into non-aggressive, medium-aggressive and high-aggressive classes. A dataset containing symbolic images and the corresponding textual comments was created to validate the proposed model. The framework employs a pre-trained VGG-16 to extract the image features and a three-layered CNN to extract the textual features in parallel. The hybrid feature set obtained by concatenating the image and the text features were optimized using the BPSO algorithm to extract the more relevant features. The proposed model with optimized features and Random Forest classifier achieves a weighted F1-Score of 0.74, an improvement of around 3% over unoptimized features.

As concerns about climate change increase, many people are calling for reductions in the use of fossil fuels and encouraging a shift to more sustainable and less polluting transportation modes. Cities and urban areas are more concerned because their population currently comprises over half of the world's population. Sustainable transportation modes such as cycling, walking, and use of public transit and electric vehicles can benefit the environment in many ways, including a reduction in toxic greenhouse gas (GHG) emissions and noise levels. In order to enhance the trend of using sustainable modes of transportation, tools, measures, and planning techniques similar to those used for vehicular transportation need to be developed. In this dissertation, we consider four problems in the context of different sustainable modes of transportation, namely, cycling, rail, public transit, and ridesharing. We develop different models to predict bike travel times for use in bike share systems (BSSs) using random forest (RF), least square boosting (LSBoost), and artificial neural network (ANN) techniques. We also use cycling Global Positioning System (GPS) data collected from 10 people (3 females and 7 males) to study cyclists' acceleration/deceleration behavior. Moreover, we develop a continuous rail transit simulator (RailSIM) intended for multi-modal energy-efficient routing applications. Finally, we propose a dynamic trip planning system that integrates ridesharing and public transit. The work done in this dissertation can help encouraging more people to move to more sustainable modes of transportation. / Doctor of Philosophy / As concerns about climate change increase, many people are calling for reductions in the use of fossil fuels and encouraging a shift to more sustainable and less polluting transportation modes. Cities and urban areas are more concerned because their population currently comprises over half of the world's population. Sustainable transportation modes such as cycling, walking, and use of public transit and electric vehicles can benefit the environment in many ways, including a reduction of toxic greenhouse gas (GHG) emissions and noise levels. In order to enhance the trend of using sustainable modes of transportation, tools, measures, and planning techniques similar to those used for vehicular transportation need to be developed. In this dissertation, we consider four problems in the context of different sustainable modes of transportation, namely, cycling, rail, public transit, and ridesharing. We develop different models to predict bike travel times in bike share systems (BSSs) using machine learning techniques. We also use cycling Global Positioning System (GPS) data collected from 10 people (3 females and 7 males) to study cyclists' acceleration/deceleration behavior. Moreover, we develop a continuous rail transit simulator (RailSIM) intended for multi-modal energy-efficient routing applications. Finally, we propose a dynamic trip planning system that integrates ridesharing and public transit. The work done in this dissertation can help encouraging more people to move to more sustainable modes of transportation.