**3.4 IUI design methods**

A poorly designed interface can bring about inconvenience to its users. Just like every other product, IUIs need to be built carefully. The need for an IUI should be obvious from the analysis of the issue, instead of making an IUI simply because it is good. First of all, it should be determined whether a device needs an IUI or not because IUIs are typically more computer-intensive than conventional user interfaces. With a user interface, if the same performance can be achieved, why bother making a more complicated and expensive IUI? The final decision whether to build an adaptive process in interfaces or not lies in weighing the expense of deployment against the enhancement of user engagement [51]. If adaptive functionality is introduced in IUIs, the cognitive processing expected by the user will be reduced whereas, installation and management of an IUI would take time and computing resources. In designing IUIs, the following iterative refining steps below are used.


*Application of Artificial Intelligence in User Interfaces Design for Cyber Security Threat Modeling DOI: http://dx.doi.org/10.5772/intechopen.96534*

one of imagination. Only getting out and seeking out fresh things and theories is the safest approach however, there are general interface architecture standards. IUI do not necessarily obey the general UI rules for example, user control, access power and consistency [52]. Other criteria, on the other hand, are best suited by IUI than by UI. The usage of natural language in IUI, for example, IUIs will speak the language of the user even more than conventional frameworks of UIs.


The following are some of the design approaches to the application of IUI for better user experience. **Table 3** shows the descriptions of various IUI design methods [53].


**Table 3.** *IUI design methods.*

#### **3.5 AI versus UI design**

The distinguishing factor between the traditional UI design and IUI design is the application of AI for better User Experience (UX). AI introduces a raw ability towards creating individualized interfaces through insights and discovery from understanding users' actions [54]. The user's expectations or intended activity on an interface is what AI feeds on to present a better user experience. The traditional UI is still relative to the design, but AI factors the user's interest to give them lasting User Experience (UX). With the traditional search engines, algorithms are rigid and yield similar search results for the same search input from different users, but AI evolves the search engines to generate smarter results tailored to a specific user. Keyboards is another rigid interface being gradually replaced by sensor-based interactions like body gestures, touchless hand gestures and sounds. Affective computing is another AI intervention, it factors human emotions, moods and expressions [55]. It's an emerging technology that enables a UI to understand and respond to human emotions in hopes of changing the rigid form of human interactions with computers.

Pertinent question must be asked when it comes to AI and UI design. What is the importance of AI to the design of User Interface? This question can only be answered with recourse to HCI. Every process in industrial design field of HCI and AI are intertwined from earliest stage in computer development with core relationship as intelligence even as at today [49]. The intelligence ushers in knowledge representation and interaction management with the application of machine learning algorithms in the design of IUI. The emergence of HCI has brought about a lot of innovations which led to technological development propelled by ubiquitous computing [56]. The idea behind HCI is making computers very easy to use and more helpful while AI is modeling human idea and exemplify those ideas into computer. These relationships bring about the creation of IUI with a high degree of usability. Therefore, AI unravels many difficulties being confronted with in User Interface design and answers many questions relating to User Experience. **Table 4** depicts HCI goals in relation to User Interface design and AI.


**Table 4.** *AI versus UI design.* *Application of Artificial Intelligence in User Interfaces Design for Cyber Security Threat Modeling DOI: http://dx.doi.org/10.5772/intechopen.96534*

The application of AI in UI design is mainly a digital practice and will visually guide the user over the product's interface. There is a brighter future for integration of AI into UI, even though there are several AI algorithms with heuristic nature but these solutions (predefined model, making more established communication as less significant than current ones and excessive accuracy) can assist in the threat modeling for cyber security [44].

## **3.6 IUI design model**

IUI design model depicts the representation of the UI and how the UX will be. The selection of design model is very important because this determines what features and services will be available in the user interface. Using a blueprint may be a major boost in the design process [57]. Some models have an overall architectural model that focuses on user modeling, whilst others propose a multi-modal input architectural model. Input is captured and later pre-processed from the keyboard, mouse, microphone, video, or probably some other input unit. Production entails case marking and other input elements that are interesting. The different modalities are fused and measured after each input modality has been evaluated.

Once all the knowledge required is accessible and revised, the framework must settle on the best choice for intervention. This is called adaptation in **Figure 6**, as some types of interface adaptation are typically selected. Assessment and adaptation sometimes take place concurrently using one inference engine for both, rendering the difference between the process of assessment and adaptation not very obvious. It is also important to produce the selected operation, which is achieved in the segment of performance production. Most IUI can be produced with this model or equipped with it. A general IUI model is shown in **Figure 6**.

#### **3.7 AI-based UI and UX prototyping tools**

AI is driving simplicity in UI and UX design, and the following are some of the top use cases.

1.**Uizard:** This tool enables for rapid prototyping of sketches. It converts handdrawn wireframes into application prototype with working code.

#### **Figure 6.**

*Intelligent user Interface model [27].*

### *Software Usability*


#### **Table 5.**

*Comparative analysis of commonly used IUI design tools.*

