Как составить описательную модель

Descriptive Models

A descriptive model describes the domain it represents in a manner that can be interpreted by humans as well as computers. It can be used for many purposes, such as those described in Chapter 2, Section 2.2.2. It can include behavioral, structural, and other descriptions that establish logical relationships about the system, such as its whole-part relationship, the interconnection between its parts, and the allocation of its behavioral elements to structural elements. Descriptive models are generally not built in a manner that directly supports simulation, animation or execution, but they can be checked for consistency and adherence to the rules of the language, and the logical relationships can be reasoned about.

The system model is a descriptive model that captures the requirements, structure, behavior, and parametric constraints associated with a system and its environment. The system model also captures inter-relationships between elements that represent its requirements, structure, behavior and parametric constraints. Because its modeling language supports various abstraction techniques, the system model also provides the ability to represent many other views of the system, such as a black-box view, white-box view, or a security view of the system. The system model can also be queried and analyzed for consistency, and serves as an integrating framework as described in Section 18.1.1.

Descriptive models

A descriptive model describes a system or other entity and its relationship to its environment. It is generally used to help specify and/or understand what the system is, what it does, and how it does it.

A geometric model or spatial model is a descriptive model that represents geometric and/or spatial relationships. Mechanical three-dimensional computer aided design (CAD) models are geometric models that include detailed information, including dimensions, tolerances, and other descriptive data such as material characteristics. A 3D representation of land topography and other features that are often presented as maps and other visualizations is also a kind of spatial model.

A logical model is a descriptive model that primarily represents logical relationships and dependencies such as functional, connectivity, and traceability relationships. Examples of logical models include a circuit design model that describes electrical components and their interconnections, and a model of system composition such as a bill-of-materials.

The system model is a logical model that is introduced in Chapter 2 Section 2.1.2. This model captures the requirements, structure, behavior, and parametric constraints associated with a system and its environment, along with the relationships between these elements. As discussed throughout this book, SysML is a modeling language used to capture the system model. SysML supports various abstraction techniques and provides the ability to represent many different views of the system, such as a black-box view, white-box view, and a security view. The system model can also be queried and analyzed for different purposes, such as providing traceability analysis, assessing the completeness of the model, and validating model correctness.

7.1.1 Delineating a problem space

A descriptive model can be as simple as a dividing up a problem space. Taken as a whole, without divisions, the problem space is … well, that’s what it is, a space—vast and uncharted, a big fuzzy cloud. However, with a little thought and organization it becomes a partitioned domain. As a partitioned domain, we are empowered to think differently about the problem space, to get inside it and see the constituent parts or processes. We can focus on certain parts of the problem space, consider how one part differs from, or relates to, another, and weigh strengths, weaknesses, advantages, or disadvantages of certain parts over others.

Johnson’s chart in Figure 7.1b is beautifully crafted. It is organized as a semi-circular spoked wheel with politics at the center. Do you like the organization? Is there a different organization that might work better? Note the symmetry between “State/Government” (left) and “Citizens/Communities” (right). Nice. “Ideas/Interests” is mirrored with “Processes.” Is that reasonable? What about “Federalism” under Institutions? That seems odd. Is federalism an institution? Perhaps Federalism, like Democracy, should be under Ideas/Interests. Enough said. The descriptive model in Figure 7.1b adds tremendous insight into politics. With it, we are empowered not only to think differently about the problem, but to think critically about it. With every tweak, refinement, and improvement we get a better model and, more importantly, we get a better understanding of the problem. That’s the power of descriptive models.

In HCI, researchers often approach problems in a similar way. Let’s have a look at a few descriptive models in HCI and examine how they are used to analyze a problem space and inform interaction design.

Model and Characteristics in Descriptive Modeling

That makes the model and its characteristics in descriptive analytics much simpler to review, understand, and use. In predictive analytics, a future event is predicted and that has to be exploited favorably. The focus is the event and, therefore, the usage is tied to the event as well. In contrast, the descriptive model output is an explanation of the data using a structured form like clustering or social network analysis. Once the output is analyzed the question of exploiting this insight becomes wide open to interpretation and innovation. Therefore, the descriptive model’s output is not expected to meet any fixed criteria. This distinction is further explained in the next section.

Decision Strategy in Descriptive Models

In descriptive models, decision strategies are still needed to address the gray area introduced by descriptive models. In case of outlier detection using clustering, the idea is to supply a large data set to a clustering algorithm and it will plot the data points and look for points that are close to form a cluster. Once the clusters are formed and their definitions have been identified, the descriptive model is ready. When a new observation comes in, it is determined which cluster it is part of or if it is close to a particular cluster. A decision strategy or the clustering software can compute that looking at the ranges of variables within each cluster and the values on the new observation. Once it is determined whether the new observation is in the cluster, the expected future behavior of the new observation will be the same as the rest of the population of the cluster, and therefore a decision can be made about decision on the new observation. On the other hand, if it lies outside a particular cluster, then it must be determined how to apply the insight of the cluster it is closest to. The Euclidean distances of “in the cluster,” “outside the cluster,” and “how far outside the cluster” are all gray areas that need a decision strategy to determine how to treat them.

Similarly, in a social network analysis, if the relationship between two customers is too strong, a certain type of decision is warranted, however, if the relationship is somewhat strong, then a different strategy is required. This again introduces a degree of gray area that requires business rules–based decision strategies. The decision strategies may be a little more volatile initially and they may go through several iterations and adjustments before getting finalized. Apart from that gray area in a model’s output, everything that has been covered in predictive decision strategies applies to descriptive analytics–based decision strategies, as the actions on the analytics’ models are still a set of DVs, their thresholds, and business decisions.

Abstract

A descriptive model of a whole enterprise consists of vision, mission statements, goals, strategies, and realizing tactics, among other things. The OMG business motivation model (BMM) provides a structure by means of which those terms can be kept apart. Hence it ensures clarity in creating an enterprise model. This chapter explains the main features of the BMM.

The BMM provides a structure for defining and developing a business plan by describing which purposes an enterprise pursues by which means. The BMM is an OMG standard and describes, on the one hand, the goals of an enterprise with a superior vision and, on the other hand, the associated implementation strategies and tactics with their superior missions. The BMM only defines the structure and properties of the BMM elements such as vision, goals, and so on, and describes their semantics. The top-most areas of BMM are: end: describes the vision of the enterprise and the goals and objectives derived thereof; means: describes which means the enterprise deploys to meet the enterprise objective; influencer: describes to which influencers the enterprise is exposed for instance, current market trends, actions of competitors, or internal influencers such as the IT infrastructure; assessment: assesses neutral influencers on goals and means used; and external information: addresses further important topics of business modeling that are not part of BMM but of other standards. The end area is divided into vision and the desired results, which in turn, are subdivided into goals and objectives. The means area is subdivided into missions, courses of action, and directives. The course of action is split into strategies and tactics, and directives into business policies and business rules. The influencers are distinguished by external and internal influencers with reference to the enterprise. The external information comprises organization units, business processes, and business rules.

Overview

Personas are descriptive models of users that embody important and realistic user characteristics that must be considered in the design process. A model of a persona is typically a one-to-three paragraph description of a user’s personality, characteristics, and use goals related to a specific design problem, often accompanied by a photo for improved realism. From an HF practitioner perspective, personas can be further enhanced through knowledge regarding aspects of the work system, such as social/organizational factors and their interaction with the physical environment.

The use of personas can be helpful in situations where iterative, rapid development is necessary, obtaining sufficient participants for more in-depth forms of user-centered or participatory design is not feasible, or when user testing may require challenging situations for the target participant group(s). For example, one study utilized personas to understand and generate interventions for patient falls, which are dangerous or difficult to recreate, and it may be challenging or harmful for those who experienced the fall to recount what happened (Hignett, Griffiths, Sands, Wolf, & Costantinou, 2013). In another study, personas were used to develop design requirements for alternative augmentative communication (AAC) devices, which help individuals who have trouble speaking and communicating. The use of personas was particularly advantageous in this case because patients requiring AAC devices typically have considerable physical and communicative limitations, making it challenging for them to actively participate in the design process for extended periods of time (Subrahmaniyan et al., 2013). Adlin and Pruitt (2010) and Cooper, Reimann, Cronin, and Noessel (2014), for example, provide detailed methodologies related to the employment of personas in design.

16.2 Oracle Data Mining Techniques

Data mining problems can be classified into two categories. In some situations you have some idea of what you are looking for—for example, you are interested in customers who are likely to buy a digital camera. This is known as directed or supervised learning. In other cases, you leave it to the mining process to find you something interesting—for example, a high incidence of accidents at a certain intersection. This is called undirected or unsupervised learning. In unsupervised learning, the data mining algorithms describe some intrinsic property or structure of data and hence are sometimes called descriptive models. On the other hand, supervised learning techniques typically use a model to predict the value or behavior of some quantity and are hence called predictive models.

Oracle Data Mining supports various techniques for mining data, each of which falls into one of these two categories.

We will look at each of these in some detail. Oracle Data Mining also provides special algorithms for mining of text and for life-sciences applications, which we will not discuss in this book.

Summary

Model for Mapping Degrees of Freedom to Dimensions

The model in Figure 3.4 is a simple tool for illustrating the mappings of degrees of freedom to dimensions for real or hypothetical input devices. The apparent shortcoming of the standard mouse as a 2D input device motivated us to design a new mouse to implement the missing DOF. As shown in the fourth column in Figure 3.4, our two-ball mouse also senses rotational motion about the Z axis. An object acquired in a graphics application with this mouse can be positioned in a single gesture both translationally along the X and Y axis and rotationally about the Z axis, thus negating the need for a “rotate tool.” See MacKenzie, Soukoreff, and Pal (1997) for details.

Figure 3.4 also shows the mappings of degrees of freedom to dimensions for other pointing devices. Column A identifies devices that sense only translation along the three axes. An example is the Owl by Pegasus Technologies, Ltd. (Tel Aviv, Israel). Column B identifies true 3D input devices, such as trackers used in virtual reality systems. Such devices sense X-, Y-, and Z-axis translation and rotation. An example is the Isotrak II by Polhemus, Inc. (Colchester, VT). Column C identifies an interesting research prototype known as the Rockin’Mouse. It looks and feels like a mouse, but it has a curved bottom and senses both translation and rotation about the X and Y axes (see Balakrishnan, Baudel, Kurtenbach, & Fitzmaurice, 1997, for details). Column D identifies a hypothetical 5-DOF input device. Because it does not sense Z-axis translation, the device is operated on the mousepad and therefore has that special mouselike appeal.