Decision Table

Software effort estimations are based on prediction properties of system with attention to develop methodologies. Many organizations follow the risk management but the risk identification techniques will differ. In this paper, we focus on two effort estimation techniques such as use case point and function point are used to estimate the effort in the software development. The decision table is used to compare these two methods to analyze which method will produce the accurate result. There are number of artificial intelligence techniques useful for training such as neural network, fuzzy logic, ant colony optimization, knowledge based system and genetic algorithm. In this paper, we used the neural network to train the decision table with the use of back propagation training algorithm and compare these two effort estimation methods (use case point and function point) with the actual effort. By using the past project data, the estimation methods are compared. Similarly risk will be evaluated by using the summary of questionnaire received from the various software developers. Based on the report, we can also mitigate the risk in the future process.

Einleitung 3
Definition 4
Entscheidungskategorien                 5
Entscheidungstechniken:                 6
➣ Pro & Contra-Listen 7
➣  Der Entscheidungsbaum 9
➣  Die Entscheidungstabelle 11
➣  Die Nutzwertanalyse 13
➣  Die Portfolio Analyse 15
➣  Die Szenario Technik 18
Fazit 19
Quellen         20

In this thesis we describe an approach and introduce a new tool, called JDeTEx, to extract decision tables from Java programs in order to give the maintenance engineer a better understanding of the control flow of the program. Decision tables are a compact, yet precise way to model the control flow of computer programs. In order to extract the decision tables the explicit and implicit control flow of Java programs needs to be analyzed by the tool. Firstly, the tool uses static byte code analysis to automatically extract information about polymorphic method calls that represent implicit control flow in Java programs.Secondly, the tool uses source code analysis to extract information about explicit control flow in the Java program. The main contribution of this thesis is an approach to instrument Java programs in order to extract decision tables from programs. We use the information extracted during byte code and source code analysis to guide the instrumentation process and show that ...

Decision tables are a widely used knowledge management tool in the decision making process. Ambiguity and conceptual reasoning difficulties arise while designing large decision tables in a collaborative environment. We introduce the notion of Semantic Decision SDT), which enhances a decision table with explicit decision semantics by annotating it properly with a domain ontology. In this paper, we focus on the SDT construction process. First, we map decision items to the ontology by building a rooted tree of decision binary facts and visualize it in a scalable manner. Formal ontological roles are used during this mapping process. Then, we commit the decision rules to the mapping results with a high level pseudo-natural language to ground their semantic. We illustrate with an SDT example from the domain of human resource management.

Our study proposes several current data mining methods to predict bankruptcy after the Sarbanes-Oxley Act (2002) using 2007-2008 U.S. data. The Sarbanes-Oxley Act (SOX) of 2002 was introduced to improve the quality of financial reporting and minimize corporate fraud in the U.S. Because of this SOX implementation, a company's financial statements are assumed to provide higher quality financial information for investors and other stakeholders. The results of our data mining approaches in our bankruptcy prediction study show that Bayesian Net method performs the best (85% overall prediction rate with 94% in AUC), followed by J48 (85% with 82% AUC), Decision Table (83.52%), and Decision Tree (82%) methods using financial and other data from the 10-K report and Compustat. These results are better than previous bankruptcy prediction studies before the SOX implementation using most current data mining approaches.

The impact of information technology (IT) on the business value of a cooperation has been an active research area for more than two decades. Although it is widely agreed that IT has a positive impact on the business values of cooperations an in-depth understanding of the underlying structures is still missing. Especially due to the huge investments in IT, there is still a need to better understand how IT influences the performance of cooperations and business values. Generally, the data collected in IT business value research to be quantitative as well as of qualitative nature. While quantitative data can be examined by classic econometric methods the analysis of qualitative data requires special methods. In the case of ordinal data DRSA -Dominance based Rough Sets Approach has been proposed. DRSA can be applied to induce rules out of a decision table containing ordinal data. This method has already successfully applied to such diverse areas like customer relationship management and satisfaction analysis, or the technical diagnostic of a fleet of vehicles besides others. In this article we apply it for the first time to the analysis of IT business value. We use ordinal data of a survey on IT management strategies of Australian firms conducted by the Australian Department of Communications, Information Technology and the Arts. The induces rules are interpreted and provide important insights into the impact of information technology on the business values of cooperations. Furthermore our study shows the potential of DRSA for information systems research where questionnaire are a widely applied technique to collect ordinal data.

Building and maintaining high quality knowledge based systems is not a trivial task. Decision tables have sometimes been recommended in this process, mainly in verification and validation. In this paper, however, it is shown how decision tables can also be used to generate, and not just to validate, knowledge bases and how the transformation process from decision tables to knowledge bases can be organized. Several options to generate rules or other knowledge representations from decision tables are described and evaluated.

An intrusion is defined as a violation of the security policy of the system, and hence, intrusion detection mainly refers to the mechanisms that are developed to detect violations of system security policy. Recently, data mining techniques have gained importance in providing the valuable information which in turn can help to enhance the decision on identifying the intrusions (attacks). In this paper; we evaluate the performance of various rule based classifiers like: JRip, RIDOR, NNge and decision table using ensemble approach in order to build an efficient network intrusion detection system. We use KDDCup'99, intrusion detection benchmark dataset (which is a part of DARPA evaluation program) for our experimentation. It can be observed from the results that the proposed approach is accurate in detecting network intrusions, provides low false positive rate, simple, reliable and faster in building an efficient network intrusion system.

Applying software defect esimation techniques and presenting this information in a compact and impactful decision table can clearly illustrate to collaborative groups how critical this position is in the overall development cycle. The Test Risk Matrix described here has proven to be a valuable addition to the management tools and approaches used in developing large scale software on several releases. Use of this matrix in development planning meetings can clarify the attendant risks and possible consequences of carrying out or bypassing specific test activities.

Most earthquake early warning systems (EEWS) developed so far are conceived as either 'regional' (network-based) or 'on-site' (stand-alone) systems. The recent implementation of nationwide, high dynamic range, dense accelerometer arrays makes now available, potentially in real time, unsaturated waveforms of moderate-to-large magnitude earthquakes recorded at very short epicentral distances (<10-20 km). This would allow for a drastic increase of the early warning lead-time, for example, the time between the alert notification and the arrival time of potentially destructive waves at a given target site. By analysing strong motion data from modern accelerograph networks in Japan, Taiwan and Italy, we propose an integrated regional/on-site early warning method, which can be used in the very first seconds after a moderate-to-large earthquake to map the most probable damaged zones. The method is based on the real-time measurement of the period (τ c ) and peak displacement (Pd) parameters at stations located at increasing distances from the earthquake epicentre. The recorded values of early warning parameters are compared to threshold values, which are set for a minimum magnitude 6 and instrumental intensity VII, according to the empirical regression analyses of strong motion data. At each recording site the alert level is assigned based on a decisional table with four alert levels defined upon critical values of the parameters Pd and τ c , which are set according to the error bounds estimated on the derived prediction equations. Given a real time, evolutionary estimation of earthquake location from first P arrivals, the method furnishes an estimation of the extent of potential damage zone as inferred from continuously updated averages of the period parameter and from mapping of the alert levels determined at the near-source accelerometer stations. The off-line application of the method to strong motion records of the Mw 6.3, 2009 Central Italy earthquake shows a very consistent match between the rapidly predicted (within a few seconds from the first recorded P wave) and observed damage zone, the latter being mapped from detailed macroseismic surveys a few days after the event. The proposed approach is suitable for Italy, where, during the last two decades, a dense network of wide dynamic-range accelerometer arrays has been deployed by the Department of Civil Protection (DPC), the Istituto Nazionale di Geofisica e Vulcanologia (INGV) and other regional research agencies.

Forecasting applications on the stock market attract much interest from researchers in the artificial intelligence field. The problem tackled in this study concerns predicting the direction of change of stock price indices, formulated in terms of binary classification. We use gene expression programming to evolve pools of binary classifiers and investigate several approaches to construct ensembles based on them. We compare the performance of the obtained classifiers with those of Naive Bayes, Support Vector Machines, Multilayer Perceptron, Decision Random Forrest. The experiments performed on real-world stock market data show that the ensembles of GEPevolved classifier models are competitive to classifiers trained by state-of-the-art machine learning methods.

Design: Current guideline document models are limited in that they reflect the specific orientation of the stakeholder who created them; thus, developers and disseminators often provide few constructs for conceptualizing recommendations, while implementers de-emphasize concepts ...

In supervised classi cation learning, one attempts to induce a classi er that correctly predicts the label of novel instances. We demonstrate that by choosing a useful subset of features for the indiscernibility relation, an induction algorithm based on simple decision table can have high prediction accuracy on arti cial and real-world datasets. We show that useful feature subsets are not necessarily maximal independent sets (relative reducts) with respect to the label, and that, in practical situations, using a subset of the relative core features may lead to superior performance.

Managers of multiple projects with overly constrained resources face difficult decisions in how to allocate resources to minimize the average delay per project or the time to complete the whole set of projects. We address the static resource-constrained multi-project scheduling problem (RCMPSP) with two lateness objectives, project lateness and portfolio lateness. In this context, past research has reported conflicting results on the performance of activity priority rule heuristics and does not provide managers with clear guidance on which rule to use in various situations. Using recently improved measures for RCMPSP characteristics, we conducted a comprehensive analysis of 20 priority rules on 12,320 test problems generated to the specifications of project-, activity-, and resource-related characteristics-including network complexity and resource distribution and contention. We found several situations in which widely advocated priority rules perform poorly. We also confirmed that portfolio managers and project managers will prefer different priority rules depending on their local or global objectives. We summarize our results in two decision tables, the practical use of which requires managers to do only a rough, qualitative characterization of their projects in terms of complexity, degree of resource contention, and resource distribution.

Intrusion detection can be considered as a classification task that attempts to classify a request to access network services as safe or malicious. Data mining techniques are being used to extract valuable information that can help in detecting intrusions. In this paper, we evaluate the performance of rule based classifiers like: JRip, RIDOR, NNge and Decision Table (DT) with Naïve Bayes (NB) along with their ensemble approach. We also propose to use the Semi-Naïve Bayesian approach (DTNB) that combines Naïve Bayes with the induction of Decision Tables in order to enhance the performance of an intrusion detection system. Experimental results show that the proposed approach is faster, reliable, and accurate with low false positive rates, which are the essential features of an efficient network intrusion detection system.

Most earthquake early warning systems (EEWS) developed so far are conceived as either `regional' (network-based) or `on-site' (stand-alone) systems. The recent implementation of nationwide, high dynamic range, dense accelerometer arrays makes now available, potentially in real time, unsaturated waveforms of moderate-to-large magnitude earthquakes recorded at very short epicentral distances (<10-20 km). This would allow for a drastic increase of the early warning lead-time, for example, the time between the alert notification and the arrival time of potentially destructive waves at a given target site. By analysing strong motion data from modern accelerograph networks in Japan, Taiwan and Italy, we propose an integrated regional/on-site early warning method, which can be used in the very first seconds after a moderate-to-large earthquake to map the most probable damaged zones. The method is based on the real-time measurement of the period (τc) and peak displacement (Pd) parameters at stations located at increasing distances from the earthquake epicentre. The recorded values of early warning parameters are compared to threshold values, which are set for a minimum magnitude 6 and instrumental intensity VII, according to the empirical regression analyses of strong motion data. At each recording site the alert level is assigned based on a decisional table with four alert levels defined upon critical values of the parameters Pd and τc, which are set according to the error bounds estimated on the derived prediction equations. Given a real time, evolutionary estimation of earthquake location from first P arrivals, the method furnishes an estimation of the extent of potential damage zone as inferred from continuously updated averages of the period parameter and from mapping of the alert levels determined at the near-source accelerometer stations. The off-line application of the method to strong motion records of the Mw 6.3, 2009 Central Italy earthquake shows a very consistent match between the rapidly predicted (within a few seconds from the first recorded P wave) and observed damage zone, the latter being mapped from detailed macroseismic surveys a few days after the event. The proposed approach is suitable for Italy, where, during the last two decades, a dense network of wide dynamic-range accelerometer arrays has been deployed by the Department of Civil Protection (DPC), the Istituto Nazionale di Geofisica e Vulcanologia (INGV) and other regional research agencies.

The do-it-yourself (DIY) culture has been continuously articulated since mid-1920s. The DIY spirit gets gradually spread because people realize that DIY activities are a way to confirm their personal creativities, outsource results, and expand their social contacts. We are motivated to design and implement a computing environment to support users to DIY their personalized Internet-of-things (IoT) applications. Before a person starts his DIY processes, it is important for him to find proper components that meet his needs. This article records our recent results concerning how to automatically discover and recommend existing components to users. The controlled fully automated ontology-assisted matching strategy (C-FOAM) is a matching strategy that contains algorithms at three different levels—string, lexical, and conceptual (or graphical). In this paper, we extend C-FOAM and embed it in a component discovery and recommendation module, which is called do-it-yourself component discoverer and recommender (DIY-CDR). DIY-CDR also uses semantic decision tables (SDT) to gather user specific decision rules and set up parameters for C-FOAM.

Today Blood donation is a global interest for world to be survival lives when people are in trouble because of natural disaster. The system provides the ability how to decide to donate the blood according to the rules for blood donation not to meet the physicians. In this system, there are three main parts to accept blood from donors when they want to donate according to the features like personal health. The application facilitates to negotiate between blood donors and patients who need to get blood seriously on page without
going to Blood Banks and waiting time in queue there

This paper proposes a comparison of the two most advanced algorithms for connected components labeling, highlighting how they perform on a soft core SoC architecture based on FPGA. In particular we test our block based connected components labeling algorithm, optimized with decision tables and decision trees. The embedded system is composed of the CMOS image sensor, FPGA, DDR SDRAM, USB controller and SPI Flash. Results highlight the importance of caching and instructions and data cache sizes for high performance image processing tasks.

The article is a study of probabilistic dependencies between attribute-defined partitions of a universe in hierarchies of probabilistic decision tables. The dependencies are expressed through two measures: the probabilistic generalization of the Pawlak's measure of the dependency between attributes and the expected certainty gain measure introduced by the author. The expected certainty gain measure reflects the subtle grades of probabilistic dependence of events. Both dependency measures are developed and it is shown how they can be extended from flat decision tables to dependencies existing in hierarchical structures of decision tables.

This study uses the variable precision rough set (VPRS) model as a tool to support group decision-making (GDM) in credit risk management. We consider the case that the classification in decision tables consisting of risk exposure (RE) may be partially erroneous, and use a variable precision factor βk to adjust the classification error. In this paper, we firstly combine VPRS and AHP to obtain the weight of condition attribute sets decided by each decision-maker (DM). Then, the integrated risk exposure (IRE) of attributes is obtained based on the three VPRS-based models. Subsequently, a new procedure of obtaining βk-stable intervals for DMk is investigated. To verify the effectiveness of these proposed methods, an illustrative example is presented. The experimental results suggest that the VPRS-based IRE have advantages in recognizing important attributes.

The problem of credit-risk evaluation is a very challenging and important financial analysis problem. Recently, researchers have found that neural networks perform very well for this complex and unstructured problem when compared to more traditional statistical approaches. A major drawback associated with the use of neural networks for decision making is their lack of explanation capability. While they can achieve a high predictive accuracy rate, the reasoning behind how they reach their decisions is not readily available. In this paper, we present the results from analyzing two real life credit-risk evaluation data sets using neural network rule extraction techniques. Clarifying the neural network decisions by explanatory rules that capture the learned knowledge embedded in the networks can help the human experts in explaining why a particular decision is made. Furthermore, we also discuss how these rules can be visualized as a decision table in a compact and intuitive graphical format. Hence, extracting rules from trained neural networks and representing these rules as a decision table may offer a viable and valuable alternative for building credit-risk evaluation expert systems.

Managers of multiple projects with overly constrained resources face difficult decisions in how to allocate resources to minimize the average delay per project or the time to complete the whole set of projects. We address the static resource-constrained multi-project scheduling problem (RCMPSP) with two lateness objectives, project lateness and portfolio lateness. In this context, past research has reported conflicting results on the performance of activity priority rule heuristics and does not provide managers with clear guidance on which rule to use in various situations. Using recently improved measures for RCMPSP characteristics, we conducted a comprehensive analysis of 20 priority rules on 12,320 test problems generated to the specifications of project-, activity-, and resource-related characteristics-including network complexity and resource distribution and contention. We found several situations in which widely-advocated priority rules perform poorly. We also confirmed that portfolio managers and project managers will prefer different priority rules depending on their local or global objectives. We summarize our results in two decision tables, the practical use of which requires managers to do only a rough, qualitative characterization of their projects in terms of complexity, degree of resource contention, and resource distribution.

This research examines which method is better for programming rule-based expert systems: IF-THEN rules or decision tables. Thirty undergraduate MIS students served as experimental subjects in an eight week study. After four weeks, subjects wrote decision tables covering cases significantly more consistently than IF-THEN rules (p<.01). After eight weeks, however, subjects wrote IF-THEN rules that were significantly more complete (p<.05) and significantly more correct (p<.01). The eighth week test also revealed that subjects preferred decision tables to IF-THEN rules (p<.05), perceiving them to be easier (p<.01). It is interesting that in spite of a significant preference for decision tables, subjects wrote significantly more accurate IF-THEN rules.

Rough Set Theory (RST) is an induction based decision-making technique, which can extract useful information from attribute-value (decision) table. This study introduces RST into pavement management system (PMS) for maintenance and rehabilitation (M&R) strategy induction. An empirical study is conducted by using the pavement distress data collected from 7 county roads by experienced pavement engineers of Taiwan Highway Bureau (THB). For each road section, the severity and coverage of existing distresses and required M&R treatment were separately recorded. The analytical database consisting of 2,348 records (2,000 records for rule induction, and 348 records for rule testing) are established to induce M&R strategies. On the basis of the testing results, total accuracy and total coverage for the induced strategies are as high as 88.7% and 84.2% respectively, which illustrates that RST certainly can reduce distress types and remove redundant records to induce the proper M&R strategies.

This paper presents work on generation of specificationdriven test cases based on quasirandom (low-discrepancy) sequences instead of pseudorandom numbers. This approach is novel in software testing. This enhanced uniformity of quasirandom sequences leads to faster generation of test cases covering all possibilities. We demonstrate by examples that quasirandom sequences can be a viable alternative to pseudorandom numbers in generating test cases. In this paper, we present a method that can generate test cases from a decision table specification more effectively via quasirandom numbers. Analysis of a simple problem in this paper shows that quasirandom sequences achieve better data than pseudorandom numbers, and have the potential to converge faster and so reduce the computational burden. The use of different quasirandom sequences for generating test cases is presented in this paper.

A service-oriented architecture (SOA) for web applications is often implemented using web services (WSs) and consists of different operations the executions of which are perceived as events. The order and time-appropriateness of occurrences of these events play a vital role for the proper functioning of a real-time SOA. This paper presents an event-based approach to modeling and testing of functional behavior of WSs by event sequence graphs (ESG). Nodes of an ESG represent events, e.g., "request" or "response", and arcs give the sequence of these events. For representing parameter values, e.g., for time-out of operation calls, ESG are augmented by decision tables. A case study carried out on a commercial web system with SOA validates the approach and analyzes its characteristic issues. The novelty of the approach stems from (i) its simplicity and lucidity in representing complex real-time web applications based on WSs in SOA, and (ii) its modeling that considers also testing and thus enables a comfortable fault management leading to a holistic view.

This paper deals with knowledge acquisition in incomplete information systems using rough set theory. The concept of similarity classes in incomplete information systems is first proposed. Two kinds of partitions, lower and upper approximations, are then formed for the mining of certain and association rules in incomplete decision tables. One type of ''optimal certain'' and two types of ''optimal association'' decision rules are generated. Two new quantitative measures, ''random certainty factor'' and ''random coverage factor'', associated with each decision rule are further proposed to explain relationships between the condition and decision parts of a rule in incomplete decision tables. The reduction of descriptors and induction of optimal rules in such tables are also examined.

This study presents ten modeling techniques within data mining process for the prediction of thermophysical properties of refrigerants (R134a, R404a, R407c and R410a). These are linear regression (LR), multi layer perception (MLP), pace regression (PR), simple linear regression (SLR), sequential minimal optimization (SMO), KStar, additive regression (AR), M5 model tree, decision table (DT), M5'Rules models. Relations depending on temperature and pressure were carried out for the determination of thermophysical properties as the specific heat capacity, viscosity, heat conduction coefficient, density of the refrigerants. Obtained model results for every refrigerant were compared and the best model was investigated. Results indicate that use of derived formulations from these techniques will facilitate design and optimize of heat exchangers which is component of especially vapor compression refrigeration system.

Iris Rose Narcisse Table Decor, is a set of 3 magnificent 100% acrylic roses from the French Brand Acrila. A modern look to any living area Marriage table decorations are an important part of the form. Still, further than one type of table will need to be decorated at a marriage and it's important that you understand everything involved. It's also important that you realize the style of the event itself, which has an impact on the decorations.

First out, the bridegroom and bachelor need to decide whether they want a head table or a squeeze table. The former is generally longer. It seats the bridegroom, the bachelor, both sets of parents, and occasionally the matrimonial party. The ultimate is a lower table which provides a more intimate setting for just the newlyweds. There's also a selection of tables for the cutlet, the food, the guests, and the gifts. Iris Rose Narcisse Table Decor : https://arkinteriors.ca/product/iris-rose-narcisse-table-decor-online/

As you can see, there are a lot of tables to embellish. This can really stretch the budget, so it's imperative to plan effects well. Generally, it's better to arrange the tables intricately, which can add aesthetic value. For case, arranging tables in a U shape is better than placing them in a straight line.

For the utmost part, white table cloths are the classic choice. They also make it easier to embellish with brightly colored plates, towels, and spectacles. You can also go with a nice centerpiece to draw attention to the table.
Understandably, some people prefer bright colors all the way around. When your table cloths are brightly covered, it makes for a veritably dramatic palette. As well, stains won't show up as fluently, which is relatively practical. After all, everybody has accidents but they need not ruin the event.

As far as redundant decorations go, you have plenitude of choices. Lots of couples like to include effects like pennants and lists. You can always subcaste your table clothes for a nice effect as well. Lace or patterns on the cloths are also nice, and drapes tend to add a sense of fineness to any event.
You can also sprinkle your table decorations. For case, numerous people choose to add a little redundant air to their tables by smattering shimmer over the table cloths and around the centerpiece. This adds a lot of redundant radiance and glamor. Confetti works the same purpose. For a more romantic mood, suppose about smattering flower petals or indeed after life leaves. However, also ocean shells make a welcome addition, If you're throwing a sand marriage.

No matter what kind of decorations you choose for your marriage table, you have the occasion to make it yours. You have the occasion to add anything you want. Rose kids, glass globules, pebbles, snow flakes- just make sure your tables reflect your personality and your marriage theme .However, also everything will clearly be beautiful, If you go that route. Including centerpieces will always add some important demanded glamor. As well, you should seriously suppose about including party favors and takeaway gifts for your guests .However, you can also include candles each around the tables for a truly romantic touch, If all additional fails.
At marriages you'll see a lot of marriage table decorations. For illustration nosegay coliseums, wreaths with and without candles, glass coliseums with floating flowers- just to name a many.

Out of all these marriage table ideas, my fave is the flower boa. A flower boa is a rather large arrangement, which sits at the front of your marriage table or event tables.
This type of marriage table decoration looks spectacular. substantially because it's relatively long. This helps to make the table look bigger, and a flower boa is great for drawing the eye in.

The name flower boa is taken from the feather boa, which a long" snake" of feathers worn around the neck. A flower boa is grounded on the same star. It's quite a long marriage table decoration with the flowers" snaked" together in a arc-suchlike shape.

The starting point of rough set theory is an information system. The information system contains data about objects of interest characterized in terms of some attributes. If we distinguish condition and decision attributes in the information system, then such a system is called a decision table. To every subset of attributes we associate a set of formulas. The decision table contains a set of decision rules. For every decision rule we associate a conditional probability which is called certainty factor. In this paper, we define the so called very positive, positive, boundary and negative regions for a decision based on the certainty factor. These regions divide the set of formulas on condition attributes according to certainty factor of decision rules. Then, some interesting relations among these divided sets of formulas are proved which are useful in decision making. Finally, a numerical example is given to show the potential application of our theoretical findings regarding select...

A three-phased framework for learning dynamic system control is presented. A genetic algorithm is employed to derive control rules encoded as decision tables. Next, the rules are automatically transformed into comprehensible form by means of inductive machine learning. Finally, a genetic algorithm is applied again to optimize the numerical parameters of the induced rules. The approach is experimentally verified on a benchmark problem of inverted pendulum control, with special emphasis on robustness and reliability. It is also shown that the proposed framework enables exploiting available domain knowledge. In this case, genetic algorithm makes qualitative control rules operational by providing interpretation of symbols in terms of numerical values.

Dominance-based Rough Set Approach (DRSA) has been introduced to deal with multiple criteria classification (also called multiple criteria sorting, or ordinal classification with monotonicity constraints), where assignments of objects may be inconsistent with respect to dominance principle. In this paper, we consider an extension of DRSA to the context of imprecise evaluations of objects on condition criteria and imprecise assignments of objects to decision classes. The imprecisions are given in the form of intervals of possible values. In order to solve the problem, we reformulate the dominance principle and introduce second-order rough approximations. The presented methodology preserves well-known properties of rough approximations, such as rough inclusion, complementarity, identity of boundaries and precisiation. Moreover, the meaning of the precisiation property is extended to the considered case. The paper presents also a way to reduce decision tables and to induce decision rules from rough approximations.