Home Research Papers1.1 received most academic consideration. While NLG is strictly

1.1 received most academic consideration. While NLG is strictly

1.1   Referring
Expression Generation

Referring expression generation (REG) is the sub-task of
Natural language generation (NLG) that received most academic consideration. While
NLG is strictly focused on transforming non-linguistic data into natural
language, REG focuses on making referring expressions that identify specified targets.
This task can be split into two sections. The content selection part recognise
which set of properties identify the proposed target and the linguistic realisation
part characterises how these properties are converted into natural language.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

There have been systematic investigations to infer how
people generate referring expressions, dating back to the 1970s (Winograd,
1972). One common approach is the Incremental Algorithm (Dale and Reiter, 1995;
Dale and Reiter, 2000) which uses logical expressions for generation. a good
deal of work in REG follows the Gricean maxims (Grice, 1975) which provide a
foundation for how people will behave in conversation. Since then there has
been a development of algorithms in the NLG community with the sole purpose of
producing a range of referring expressions.

1.1.1    GIVE

The major challenge on Generating Instructions in Virtual
Environments (GIVE) is an interesting approach to the incredibly tough problem
of NLG systems evaluation.  For NLG
systems, GIVE represents an end-to-end-evaluation. The software allows each
player in a 3D game world to be connected with one another with the aid of an
NLG system over the Internet. In the GIVE scenario, for instance, users are
involved in a treasure hunt in a virtual environment they are not familiar
with. The complete representation of the virtual environment is provided by the
computer. What the NLG system aims to achieve here is to provide
natural-language instructions (in real time) that will help users complete
their task. It has to do with research opportunities in sentence planning,
realization, text planning, and situated communication.

One fascinating aspect of situating the problem of
generation in a virtual setting is the fact that relational and spatial
expressions play a more prominent role than in other NLG tasks. Aside from NLG,
GIVE could be beneficial when used as a testbed to improve the components of
dialogue systems. GIVE has proven to provide consistent results that are also
more comprehensive than those obtained from a traditional lab-based evaluation.

1.1.2    RefCOCOg

This is a dataset collected from in a non-interactive
setting on Amazon Mechanical Turk. This data is comprised of 85,474 referring
expressions for 54,822 in 26,711 images. Selected images contained between 2
and 4 objects of the same category of subject.







1.2   User

User Experience (UX) concerns user’s feelings and mentalities
about using specific systems or services. It includes meaningful, practical and
often experiential, and important parts of human–computer interaction.
Moreover, it includes user’s perceptions of framework aspects like i.e.
usability, efficiency and utility.

In order for there to be a valuable and purposeful UX, information
has to be:

Useful: Content should be original and fulfil
a need

Usable: Site must be easy to use

Desirable: Image, identity, brand, and other
design elements are used to evoke emotion and appreciation

Findable: Content needs to be navigable and
locatable onsite and offsite

Accessible: Content needs to be accessible to
people with disabilities

Credible: Users must trust and believe in
presented content

User experience (UX) concentrates on having a profound
understanding of users, what they require, what they esteem, their abilities,
and limitations. Likewise, it considers the business objectives and goals of
the team managing the project. User experience best practices encourage to
improve quality of the user’s interaction with and impression of the product or
services. UX is dynamic as it is continually changing with time because of
circumstances we use the new and upgraded systems are also changing.



1.3   PHP Scripting

PHP (Hypertext Pre-Processor) is a broadly used open source and general-purpose
scripting language that is ideally suited for web development (php.net. N.p.,
n.d., 2017). Unlike client side languages i.e. JavaScript, PHP code is executed
on the server, where HTML is generated and then sent to the client. The output
of running that script is displayed to the client but what client don’t know,
is the underlying code. What characterise PHP is that it’s fairly
straightforward for a newcomer, however offers numerous advanced features for
professional programmers as well.

The main focus of PHP is on server-side scripting, so you can
achieve anything that can be achieved with any other CGI program, i.e.
receiving and sending cookies, dynamic page generation, or data collection. Two
things are needed to make these things work: a web server and PHP parser
(server module or CGI), web browser. You have to run the web server, with
working PHP installation. Output of PHP program is accessible with any web
browser, while viewing the page through the web server. PHP can be easily used
on every major operating system, whether it’s MS Windows, Mac OS X or Linux.
PHP also offers support for almost all of today’s web servers, including
Microsoft’s IIS, Apache or basically any server able to utilise FastCGI PHP



Beginning web development with PHP, you have the flexibility of
choosing OS and the web server. You also have the option of using
object-oriented or procedural programming, or you can combine both methods. With
PHP, you are not restricted to only HTML output. PHP allows outputting PDF files,
images and even Flash animations. It’s also possible to output any text easily,
such as XHTML or even other file types. Those files can be auto-generated by
PHP and saved in the file system, rather than just printing them out and
creating a server-side cache for dynamic content.


I'm Alfred!

Would you like to get a custom essay? How about receiving a customized one?

Check it out