Test cost and misclassification cost trade-off using reframing
نویسندگان
چکیده
Many solutions to cost-sensitive classification (and regression) rely on some or all of the following assumptions: we have complete knowledge about the cost context at training time, we can easily re-train whenever the cost context changes, and we have technique-specific methods (such as cost-sensitive decision trees) that can take advantage of that information. In this paper we address the problem of selecting models and minimising joint cost (integrating both misclassification cost and test costs) without any of the above assumptions. We introduce methods and plots (such as the so-called JROC plots) that can work with any off-the-shelf predictive technique, including ensembles, such that we reframe the model to use the appropriate subset of attributes (the feature configuration) during deployment time. In other words, models are trained with the available attributes (once and for all) and then deployed by setting missing values on the attributes that are deemed ineffective for reducing the joint cost. As the number of feature configuration combinations grows exponentially with the number of features we introduce quadratic methods that are able to approximate the optimal configuration and model choices, as shown by the experimental results.
منابع مشابه
A Multi-Mode Resource-Constrained Optimization of Time-Cost Trade-off Problems in Project Scheduling Using a Genetic Algorithm
In this paper, we present a genetic algorithm (GA) for optimization of a multi-mode resource constrained time cost trade off (MRCTCT) problem. The proposed GA, each activity has several operational modes and each mode identifies a possible executive time and cost of the activity. Beyond earlier studies on time-cost trade-off problem, in MRCTCT problem, resource requirements of each execution mo...
متن کاملA Hybrid Method Based on Fuzzy AHP and VIKOR for the Discrete Time-Cost-Quality Trade-off Problem
Time, cost and quality are considered as the main components in managing each project. Previous researches have mainly focused on the time-cost trade-off problems. Recently quality is considered as the most important factor in project�s success, which is influenced by time acceleration that is the less time is spent the more success is gained. In time-cost-quality trade-off problems, each act...
متن کاملA New Heuristic Algorithm for Time-cost Trade-off Problem Taking into Account Monetary Value
Time-cost trade-off is one of the most important subjects in project management and of interest to contractors. The goal of time-cost trade-off is sensivity analysis of project costs to changes in activity duration in order to obtain the best combination of activity duration decrease, in a way that the sum of project costs is minimized. In the heuristics presented in this area, time crashing is...
متن کاملAN EXTENSION TO STOCHASTIC TIME-COST TRADE-OFF PROBLEM OPTIMIZATION WITH DISCOUNTED CASH FLOW
In this paper, an efficient multi-objective model is proposed to solve time-cost trade off problem considering cash flows. The proposed multi-objective meta-heuristic is based on Ant colony optimization and is called Non Dominated Archiving Ant Colony Optimization (NAACO). The significant feature of this work is consideration of uncertainties in time, cost and more importantly interest rate. A ...
متن کاملA New Heuristic Algorithm for Time-cost Trade-off Problem Taking into Account Monetary Value
Time-cost trade-off is one of the most important subjects in project management and of interest to contractors. The goal of time-cost trade-off is sensivity analysis of project costs to changes in activity duration in order to obtain the best combination of activity duration decrease, in a way that the sum of project costs is minimized. In the heuristics presented in this area, time crashing is...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1305.7111 شماره
صفحات -
تاریخ انتشار 2013