IS

Mookerjee, Vijay S.

Research Interests (53)

List of Topics (53)

Topic Weight Topic Terms
2.048 set approach algorithm optimal used develop results use simulation experiments algorithms demonstrate proposed optimization present
0.862 firms firm financial services firm's size examine new based result level including results industry important
0.814 policy movie demand features region effort second threshold release paid number regions analyze period respect
0.811 expert systems knowledge knowledge-based human intelligent experts paper problem acquisition base used expertise intelligence domain
0.625 consumer consumers model optimal welfare price market pricing equilibrium surplus different higher results strategy quality
0.589 errors error construction testing spreadsheet recovery phase spreadsheets number failures inspection better studies modules rate
0.576 security threat information users detection coping configuration avoidance response firm malicious attack intrusion appraisal countermeasures
0.499 customer customers crm relationship study loyalty marketing management profitability service offer retention it-enabled web-based interactions
0.484 increased increase number response emergency monitoring warning study reduce messages using reduced decreased reduction decrease
0.457 programming program programmers pair programs pairs software development problem time language application productivity best nominal
0.377 use habit input automatic features modification different cognition rules account continuing underlying genre emotion way
0.371 source open software oss development developers projects developer proprietary community success openness impact paper project
0.333 process problem method technique experts using formation identification implicit analysis common proactive input improvements identify
0.332 costs cost switching reduce transaction increase benefits time economic production transactions savings reduction impact services
0.325 pricing services levels level on-demand different demand capacity discrimination mechanism schemes conditions traffic paper resource
0.308 piracy goods digital property intellectual rights protection presence legal consumption music consumers enforcement publisher pirate
0.290 project projects development management isd results process team developed managers teams software stakeholders successful complex
0.285 content providers sharing incentive delivery provider net incentives internet service neutrality broadband allow capacity congestion
0.281 internal external audit auditing results sources closure auditors study control bridging appears integrity manager effectiveness
0.266 results study research information studies relationship size variables previous variable examining dependent increases empirical variance
0.258 information types different type sources analysis develop used behavior specific conditions consider improve using alternative
0.230 learning model optimal rate hand domain effort increasing curve result experts explicit strategies estimate acquire
0.222 methods information systems approach using method requirements used use developed effective develop determining research determine
0.222 recommendations recommender systems preferences recommendation rating ratings preference improve users frame contextual using frames sensemaking
0.218 software development product functionality period upgrade sampling examines extent suggests factors considered useful uncertainty previous
0.199 problem problems solution solving problem-solving solutions reasoning heuristic theorizing rules solve general generating complex example
0.186 high low level levels increase associated related characterized terms study focus weak hand choose general
0.185 online consumers consumer product purchase shopping e-commerce products commerce website electronic results study behavior experience
0.185 use question opportunities particular identify information grammars researchers shown conceptual ontological given facilitate new little
0.182 users user new resistance likely benefits potential perspective status actual behavior recognition propose user's social
0.170 effect impact affect results positive effects direct findings influence important positively model data suggest test
0.166 strategies strategy based effort paper different findings approach suggest useful choice specific attributes explain effective
0.165 product products quality used characteristics examines role provide goods customization provides offer core sell key
0.165 data classification statistical regression mining models neural methods using analysis techniques performance predictive networks accuracy
0.164 advertising search online sponsored keywords sales revenue advertisers ads keyword organic advertisements selection click targeting
0.160 knowledge sharing contribution practice electronic expertise individuals repositories management technical repository knowledge-sharing shared contributors novelty
0.156 states united employment compensation labor workers paper work extent findings increasing implications concerns relationship managerial
0.152 performance results study impact research influence effects data higher efficiency effect significantly findings impacts empirical
0.147 structure integration complex business enhancement effects access extent analyzing volatile capture requires occurs pattern enables
0.143 contract contracts incentives incentive outsourcing hazard moral contracting agency contractual asymmetry incomplete set cost client
0.142 model models process analysis paper management support used environment decision provides based develop use using
0.138 coordination mechanisms work contingencies boundaries temporal coordinating vertical associated activities different coordinate suggests dispersed coordinated
0.137 value business benefits technology based economic creation related intangible cocreation assessing financial improved key economics
0.137 time use size second appears form larger benefits combined studies reasons selected underlying appear various
0.135 office document documents retrieval automation word concept clustering text based automated created individual functions major
0.130 quality different servqual service high-quality difference used quantity importance use measure framework impact assurance better
0.125 behavior behaviors behavioral study individuals affect model outcomes psychological individual responses negative influence explain hypotheses
0.116 software vendors vendor saas patch cloud release model vulnerabilities time patching overall quality delivery software-as-a-service
0.115 performance firm measures metrics value relationship firms results objective relationships firm's organizational traffic measure market
0.111 phase study analysis business early large types phases support provided development practice effectively genres associated
0.110 decision making decisions decision-making makers use quality improve performance managers process better results time managerial
0.102 e-commerce value returns initiatives market study announcements stock event abnormal companies significant growth positive using
0.100 participation activities different roles projects examined outcomes level benefits conditions key importance isd suggest situations

Coauthor Network

■ Focal Researcher     ■ Coauthors of Focal Researcher (1st degree)     ■ Coauthors of Coauthors (2nd degree)

Note: click on a node to go to a researcher's profile page. Drag a node to reallocate. Number on the edge is the number of co-authorships.

List of Coauthors (26)

Kumar, Subodha 4 Ji, Yonghua 3 Johar, Monica S. 3 Mannino, Michael V. 3
Sarkar, Sumit 2 Tan, Yong 2 Bensoussan, Alain 1 Chiang, I. Robert 1
Chen, Hongyu 1 Dos Santos, Brian L. 1 Dawande, Milind 1 Gilson, Robert 1
Ghoshal, Abhijeet 1 Jiang, Zhengrui 1 Jabr, Wael 1 Koushik, Murlidhar V. 1
Kumar, Nanda 1 Liu, Dengpan 1 Mookerjee, Radha 1 Mehra, Amit 1
Mookerjee, Radha V. 1 Radhakrishnan, Suresh 1 Sethi, Suresh P. 1 Santos, Brian L. Dos 1
Singh, Param Vir 1 Zheng, Zhiqiang (Eric) 1

Article Keywords (89)

inductive expert systems 3 Nash equilibrium 2 optimal control theory 2 advertising 1
business value of IT 1 Coordination policy 1 controlled scrambling 1 Case Based Systems 1
Case Retrieval Algorithms 1 Cost Reduction 1 concurrent development and debugging 1 content piracy 1
content provision and distribution 1 cohesion 1 differential game 1 delivery delay 1
demand endogeneity 1 demand variability 1 duopoly pricing 1 dynamic optimization 1
economic expert system design 1 economics of machine learning 1 Economic model 1 expert systems 1
extreme programming 1 event study 1 employment contracts 1 electronic retailing 1
financial market evaluation 1 feedback 1 genetic algorithms 1 heuristics 1
hacker learning 1 Human capital 1 information costs and benefits 1 input data noise 1
Information Costs 1 incremental development 1 input distortion 1 integer programming 1
information technology industry 1 IT and firm performance 1 IT value 1 IT capacity 1
IT security 1 Joint Versus Separate Optimization 1 macroeconomic news 1 monitoring and profiling 1
noise handling 1 optimal software development 1 optimal security management 1 open source software development 1
open source software 1 optimal control 1 outsourcing 1 optimization 1
online competition 1 performance stability 1 pair programming 1 P2P networks 1
project success 1 pro-social behavior 1 quality-driven integration policy 1 Risk Aversion in Expert Systems 1
reneging 1 recognition mechanism 1 recommender systems 1 Software construction 1
software project management 1 sequential information gathering 1 software development methodology 1 security shocks 1
stock price volatility 1 Social networks 1 skill development incentives 1 software upgrades 1
Team size 1 Tradeoffs 1 team coordination 1 team composition 1
training 1 text mining 1 upgrade design effort 1 upgrade strategy 1
User support forum 1 variance reduction 1 Value-Based System Design 1 variable attack rates 1
web-based personalization 1

Articles (20)

When Being Hot Is Not Cool: Monitoring Hot Lists for Information Security (Information Systems Research, 2016)
Authors: Abstract:
    We study operational and managerial problems arising in the context of security monitoring where sessions, rather than raw individual events, are monitored to prevent attacks. The objective of the monitoring problem is to maximize the benefit of monitoring minus the monitoring cost. The key trade-off in our model is that as more sessions are monitored, the attack costs should decrease. However, the monitoring cost would likely increase with the number of sessions being monitored. A key step in solving the problem is to derive the probability density of a system with n sessions being monitored with a session's age measured as the time elapsed since it last generated a suspicious event. We next optimize the number of sessions monitored by trading off the attack cost saved with the cost of monitoring. A profiling step is added prior to monitoring and a resulting two-dimensional optimization problem is studied. Through numerical simulation, we find that a simple size-based policy is quite robust for a very reasonable range of values and, under typical situations, performs almost as well as the two more sophisticated policies do. Also, we find that adopting a simplified policy without using the option of managing sessions using age threshold can greatly increase the ease of finding an optimal solution, and reduce operational overhead with little performance loss compared with a policy using such an option. The insights gained from the mechanics of profiling and monitoring are leveraged to suggest a socially optimal contract for outsourcing these activities in a reward-based contract. We also study penalty-based contracts. Such contracts (specifically, when the penalty is levied as a percentage of the monthly service fee) do not achieve the social optimum. We show how an appropriate penalty coefficient can be chosen to implement a socially optimal penalty-based contract. In addition, we provide a high-level comparison between reward- and penalty-based contracts. In a penalty-based contract, the setting of the fixed payment can be challenging because it requires additional knowledge of the total expected malicious event rate, which needs to be observed through a period of no monitoring.
Impact of Recommender System on Competition Between Personalizing and Non-Personalizing Firms (Journal of Management Information Systems, 2015)
Authors: Abstract:
    How do recommender systems affect prices and profits of firms under competition? To explore this question, we model the strategic behavior of customers who make repeated purchases at two competing firms: one that provides personalized recommendations and another that does not. When a customer intends to purchase a product, she obtains recommendations from the personalizing firm and uses this recommendation to eventually purchase from one of the firms. The personalizing firm profiles the customer (based on past purchases) to recommend products. Hence, if a customer purchases less frequently from the personalizing firm, the recommendations made to her become less relevant. While considering the impact on the quality of recommendations received, a customer must balance two opposing forces: (1) the lower price charged by the non-personalizing firm, and (2) an additional fit cost incurred when purchasing from the non-personalizing firm and the increased cost due to recommendations of reduced quality in the future. An outcome of the analysis is that the customers should distribute their purchases across both firms to maximize surplus over a planning horizon. Anticipating this response, the firms simultaneously choose prices. We study the sensitivity of the equilibrium prices and profits of the firms with respect to the effectiveness of the recommender system and the profile deterioration rate. We also analyze some interesting variants of the base model in order to study how its key results could be influenced. One of the key takeaways of this research is that the recommender system can influence the price and profit of not only the personalizing firm but also the non-personalizing firm. > >
Selling vs. Profiling: Optimizing the Offer Set in Web-Based Personalization (Information Systems Research, 2014)
Authors: Abstract:
    We study the problem of optimally choosing the composition of the offer set for firms engaging in web-based personalization. A firm can offer items or links that are targeted for immediate sales based on what is already known about a customer's profile. Alternatively, the firm can offer items directed at learning a customer's preferences. This, in turn, can help the firm make improved recommendations for the remainder of the engagement period with the customer. An important decision problem faced by a profit maximizing firm is what proportion of the offer set should be targeted toward immediate sales and what proportion toward learning the customer's profile. We study the problem as an optimal control model, and characterize the solution. Our findings can help firms decide how to vary the size and composition of the offer set during the course of a customer's engagement period with the firm. The benefits of the proposed approach are illustrated for different patterns of engagement, including the length of the engagement period, uncertainty in the length of the period, and the frequency of the customer's visits to the firm. We also study the scenario where the firm optimizes the size of the offer set during the planning horizon. One of the most important insights of this study is that frequent visits to the firm's website are extremely important for an e-tailing firm even though the customer may not always buy products during these visits.
Leveraging Philanthropic Behavior for Customer Support: The Case of User Support Forums (MIS Quarterly, 2014)
Authors: Abstract:
    Online user forums for technical support are being widely adopted by IT firms to supplement traditional customer support channels. Customers benefit from having an additional means of product support, while firms benefit by lowering the costs of supporting a large customer base. Typically these forums are populated with content generated by users, consisting of questioners (solution seekers) and solvers (solution providers). While questioners can be expected to keep returning as long as they can find answers, firms must employ different means in order to recognize and encourage the contributions of solvers. We identify and compare the impact of two widely adopted recognition mechanisms on the philanthropic behavior of solvers. In the first mechanism, feedback-based recognition, solver contribution is evaluated by questioners. In the second mechanism, quantity-based recognition, all contributions are weighted equally regardless of questioner feedback. We draw on the pro-social behavior literature to identify four drivers of solver contribution: (1) peer recognition, (2) image motivation, (3) social comparison, and (4) social exposure. We show that the choice of recognition mechanism strongly influences a solver’s problem-solving behavior, highlighting the importance of the firm’s decision in this regard. We address issues of solvers self-selecting a type of recognition mechanism by using propensity score analysis in order to show that solver behavior is a result of forum conditioning. We also study the impact of the recognition mechanism on forum quality and the effectiveness of support to draw comparative analytics.
Are New IT-Enabled Investment Opportunities Diminishing for Firms? (Information Systems Research, 2012)
Authors: Abstract:
    Today, few firms could survive for very long without their computer systems. IT has permeated every corner of firms. Firms have reached the current state in their use of IT because IT has provided myriad opportunities for firms to improve performance and, firms have availed themselves of these opportunities. Some have argued, however, that the opportunities for firms to improve their performance through new uses of IT have been declining. Are the opportunities to use IT to improve firm performance diminishing? We sought to answer this question. In this study, we develop a theory and explain the logic behind our empirical analysis; an analysis that employs a different type of event study. Using the volatility of firms' stock prices to news signaling a change in economic conditions, we compare the stock price behavior of firms in the IT industry to firms in the utility and transportation and freight industries. Our analysis of the IT industry as a whole indicates that the opportunities for firms to use IT to improve their performance are not diminishing. However, there are sectors within the IT industry that no longer provide value-enhancing opportunities for firms. We also find that IT products that provided opportunities for firms to create value at one point in time, later become necessities for staying in business. Our results support the key assumption in our work.
Advertising Strategies in Electronic Retailing: A Differential Games Approach. (Information Systems Research, 2012)
Authors: Abstract:
    We consider advertising problems under an information technology (IT) capacity constraint encountered by electronic retailers in a duopolistic setting. There is a considerable amount of literature on advertising games between firms, yet introducing an IT capacity constraint fundamentally changes this problem. In the presence of information processing constraints, although advertising may still cause a customer to switch, it may not result in a sale, i.e., the customer may be lost by both firms. This situation could occur when customers have a limited tolerance for processing delays and leave the website of a firm because of slow response. In such situations, attracting more traffic to a firm's site (by increasing advertising expenditure) may not generate enough additional revenue to warrant this expenditure. We use a differential game formulation to obtain closedform solutions for the advertising effort over time in the presence of IT capacity constraints. Based on these solutions, we present several useful managerial insights.
Content Provision Strategies in the Presence of Content Piracy. (Information Systems Research, 2012)
Authors: Abstract:
    We consider a publisher that earns advertising revenue while providing content to serve a heterogeneous population of consumers. The consumers derive benefit from consuming content but suffer from delivery delays. A publisher's content provision strategy comprises two decisions: (a) the content quality (affecting consumption benefit) and (b) the content distribution delay (affecting consumption cost). The focus here is on how a publisher should choose the content provision strategy in the presence of a content pirate such as a peer-to-peer (P2P) network. Our study sheds light on how a publisher could leverage a pirate's presence to increase profits, even though the pirate essentially encroaches on the demand for the publisher's content. We find that a publisher should sometimes decrease the delivery speed but increase quality in the presence of a pirate (a quality focused strategy). At other times, a distribution focused strategy is better; namely, increase delivery speed, but lower quality. In most cases, however, we show that the publisher should improve at least one dimension of content provision (quality or delay) in the presence of a pirate.
HUMAN CAPITAL DEVELOPMENT FOR PROGRAMMERS USING OPEN SOURCE SOFTWARE. (MIS Quarterly, 2012)
Authors: Abstract:
    A firm can upgrade relevant skills of its programmers by ensuring their participation in carefully chosen open source projects. Highly skilled programmers are more valuable for the firm but participating in open source projects reduces the time they spend doing the firm's projects. This tradeoff determines the optimal extent of programmer participation in open source for the firm. The extent of open source participation may also be influenced by the minimum compensation that must be paid to hire a programmer in the labor market. This is because providing better skills is a way of compensating the programmers by improving their future market value. Hence the firm may want to increase open source participation to keep direct wage payments in check.We develop an analytical model based on optimal control theory to characterize the employment contract that features the best mix of open source participation and wage payments. We also find that the firm benefits more from the presence of open source in a tight labor market (i.e., when programmers have good options besides the employment offered by the firm). On the other hand, programmers are compensated better in the presence of open source opportunities when they have few outside options. This benefit is more for less skilled programmers.
When Hackers Talk: Managing Information Security Under Variable Attack Rates and Knowledge Dissemination. (Information Systems Research, 2011)
Authors: Abstract:
    This paper analyzes interactions between a firm that seeks to discriminate between normal users and hackers that try to penetrate and compromise the firm's information assets. We develop an analytical model in which a variety of factors are balanced to best manage the detection component within information security management. The approach not only considers conventional factors such as detection rate and false-positive rate, but also factors associated with hacker behavior that occur in response to improvements in the detection system made by the firm. Detection can be improved by increasing the system's discrimination ability (i.e., the ability to distinguish between attacks and normal usage) through the application of maintenance effort. The discrimination ability deteriorates over time due to changes in the environment. Also, there is the possibility of sudden shocks that can sharply degrade the discrimination ability. The firm's cost increases as hackers become more knowledgeable by disseminating security knowledge within the hacker population. The problem is solved to reveal the presence of a steady-state solution in which the level of system discrimination ability and maintenance effort are held constant. We find an interesting result where, under certain conditions, hackers do not benefit from disseminating security knowledge among one another. In other situations, we find that hackers benefit because the firm must lower its detection rate in the presence of knowledge dissemination. Other insights into managing detection systems are provided. For example, the presence of security shocks can increase or decrease the optimal discrimination level as compared to the optimal level without shocks.
Managing the Versions of a Software Product Under Variable and Endogenous Demand (Information Systems Research, 2011)
Authors: Abstract:
    Software product versioning (i.e., upgrading the product after its initial release) is a widely adopted practice followed by leading software providers such as Microsoft, Oracle, and IBM. Unlike conventional durable goods, software products are relatively easy to upgrade, making upgrades a strategic consideration in commercial software production. We consider a two-period model with a monopoly software provider who develops and releases a software product to the market. Unlike previous research, we consider demand variability and endogeneity to determine the functionality of the software in the first and second periods. Demand endogeneity is the impact of the word-of-mouth effect that positively relates the features in the initial release of the product to its demand in the second period. We also determine the design effort that should be spent in the first period to prepare for upgrading the product in the second period—upgrade design effort—to tap into the possible future demand. Results show that the upgrade design effort can be lower or higher when there is more market demand uncertainty. We also show that the features of the product in its initial release and upgrade design effort can be complements as well as substitutes, depending on the strength of the word-of-mouth effect. The results in this paper provide insights into how demand-side factors (market demand variability or demand endogeneity) can influence supply-side decisions (initial features and upgrade design effort). A key insight of the analysis is that a high word-of-mouth effect helps manage the product in the face of demand variability.
NETWORK EFFECTS: THE INFLUENCE OF STRUCTURAL CAPITAL ON OPEN SOURCE PROJECT SUCCESS. (MIS Quarterly, 2011)
Authors: Abstract:
    What determines the success of open source projects? In this study, we investigate the impact of network social capital on open source project success. We define network social capital as the benefits open source developers secure from their membership in developer collaboration networks. We focus on one specific type of success as measured by the rate of knowledge creation in an open source project. Specific hypotheses are developed and tested using a longitudinal panel of 2,378 projects hosted at SourceForge. We find that network social capital is not equally accessible to or appropriated by all projects. Our main results are as follows. First, projects with greater internal cohesion (that is, cohesion among the project members) are more successful. Second, external cohesion (that is, cohesion among the external contacts of a project) has an inverse U-shaped relationship with the project’s success; moderate levels of external cohesion are best for a project’s success rather than very low or very high levels. Third, the technological diversity of the external network of a project also has the greatest benefit when it is neither too low nor too high. Fourth, the number of direct and indirect external contacts positively affects a project’s success such that the effect of the number of direct contacts is moderated by the number of indirect contacts. These results are robust to several control variables and alternate model specifications. Several theoretical and managerial implications are provided.
A Comparison of Pair Versus Solo Programming Under Different Objectives: An Analytical Approach. (Information Systems Research, 2008)
Authors: Abstract:
    This study compares the performances of pair development (an approach in which a pair of developers jointly work on the same piece of code), solo development, and mixed development under two separate objectives: effort minimization and time minimization. To this end, we develop analytical models to optimize module-developer assignments in each of these approaches. These models are shown to be strongly NP-hard and solved using a genetic algorithm. The solo and pair development approaches are compared for a variety of problem instances to highlight project characteristics that favor one of the two practices. We also propose a simple criterion that can reliably recommend the appropriate approach for a given problem instance. Typically, for efficient knowledge sharing between developers or for highly connected systems, the pair programming approach is preferable. Also, the pair approach is better at leveraging expertise by pairing experts with less skilled partners. Solo programming is usually desirable if the system is large or the effort needed either to form a pair or to code efficiently in pairs is high. Solo programming is also appropriate for projects with a tight deadline, whereas the reverse is true for projects with a lenient deadline. The mixed approach (i.e., an approach where both the solo and pair practices are used in the same project) is only indicated when the system consists of groups of modules that are sufficiently different from one another.
Lying on the Web: Implications for Expert Systems Redesign. (Information Systems Research, 2005)
Authors: Abstract:
    We consider a new variety of sequential information gathering problems that are applicable for Web-based applications in which data provided as input may be distorted by the system user, such as an applicant for a credit card. We propose two methods to compensate for input distortion. The first method, termed knowledge base modification, considers redesigning the knowledge base of an expert system to best account for distortion in the input provided by the user. The second method, termed input modification, modifies the input directly to account for distortion and uses the modified input in the existing (unmodified) knowledge base of the system. These methods are compared with an approach where input noise is ignored. Experimental results indicate that both types of modification substantially improve the accuracy of recommendations, with knowledge base modification outperforming input modification in most cases. Knowledge base modification is, however, more computationally intensive than input modification. Therefore, when computational resources are adequate, the knowledge base modification approach is preferred; when such resources are very limited, input modification may be the only viable alternative.
Optimal Software Development: A Control Theoretic Approach. (Information Systems Research, 2005)
Authors: Abstract:
    We study the problem of optimally allocating effort between software construction and debugging. As construction proceeds, new errors are introduced into the system. The objective is to deliver a system of the highest possible quality (fewest number of errors) subject to the constraint that N system modules are constructed in a specified duration T. If errors are not corrected during construction, then further construction can produce errors at a faster rate. To curb the growth of errors, some of the effort must be taken away from construction and assigned to testing and debugging. A key finding of this model is that the practice of alternating between pure construction and pure debugging is suboptimal. Instead, it is desirable to concurrently construct and debug the system. We extend the above model to integrate decisions traditionally considered "external" such as the time to release the product to the market with those that are typically treated as "internal" such as the division of effort between construction and debugging. Results show that integrating these decisions can yield significant reduction in the overall cost. Also, when competitive forces are strong, it may be better to release a product early (with more errors) than late (with fewer errors). Thus, underestimating the cost of errors in the product may be better than overestimating the cost.
A Fault Threshold Policy to Manage Software Development Projects. (Information Systems Research, 2004)
Authors: Abstract:
    This paper presents a project management policy in which the appearance of software faults during system construction is used to determine the timing of system integration activities (e.g., team meetings, analyzing modules for interface inconsistencies, system fault correction, and so on). System integration is performed only if a threshold fault count has been exceeded; otherwise, module development is allowed to continue. We derive an expression for calculating fault thresholds and analyze the policy to reveal the presence of three operating regions: (1) a region in which development should continue with no system integration, (2) a region in which system integration occurs if a threshold fault count has been exceeded, and (3) a region in which system integration should always take place. Analytical and numerical results demonstrate how the fault thresholds change with system complexity, team skill, development environment, and project schedule. We also show how learning that occurs during each round of system integration leads to less frequent integration in the future, and lower total construction effort. Simulation experiments reveal that the fault threshold policy can be applied even if several homogeneity assumptions in the model are relaxed, allowing for differences in the propensity among modules to accumulate faults and the effort needed to correct these faults. Finally, the fault threshold policy outperforms a fixed-release policy in which system integration occurs whenever a fixed number of modules has been released.
Mean-Risk Trade-Offs in Inductive Expert Systems. (Information Systems Research, 2000)
Authors: Abstract:
    Notably absent in previous research on inductive expert systems is the study of mean-risk trade-offs. Such trade-offs may be significant when there are asymmetries such as unequal classification costs, and uncertainties in classification and information acquisition costs. The objective of this research is to develop models to evaluate mean-risk trade-offs in value-based inductive approaches. We develop a combined mean-risk measure and incorporate it into the Risk-Based induction algorithm. The mean-risk measure has desirable theoretical properties (consistency and separability) and is supported by empirical results on decision making under risk. Simulation results using the Risk-Based algorithm demonstrate: (i) an order of magnitude performance difference between mean-based and risk-based algorithms and (ii) an increase in the performance difference between these algorithms as either risk aversion, uncertainty, or asymmetry increases given modest thresholds of the other two factors.
Redesigning Case Retrieval to Reduce Information Acquisition Costs. (Information Systems Research, 1997)
Authors: Abstract:
    Retrieval of a set of cases similar to a new case is a problem common to a number of machine learning approaches such as nearest neighbor algorithms, conceptual clustering, and case based reasoning. A limitation of most case retrieval algorithms is their lack of attention to information acquisition costs. When information acquisition costs are considered, cost reduction is hampered by the practice of separating concept formation and retrieval strategy formation. To demonstrate the above claim, we examine two approaches. The first approach separates concept formation and retrieval strategy formation. To form a retrieval strategy in this approach, we develop the CR<sub>1c</sub> (case retrieval loss criterion) algorithm that selects attributes in ascending order of expected loss. The second approach jointly optimizes concept formation and retrieval strategy formation using a cost based variant of the ID3 algorithm (ID3<sub>c</sub>). ID3<sub>c</sub> builds a decision free wherein attributes are selected using entropy reduction per unit information acquisition cost. Experiments with four data sets are described in which algorithm, attribute cost coefficient of variation, and matching threshold are factors. The experimental results demonstrate that (i) jointly optimizing concept formation and retrieval strategy formation has substantial benefits, and (ii) using cost considerations can significantly reduce information acquisition costs, even if concept formation and retrieval strategy formation are separated.
Modeling Coordination in Software Construction: An Analytical Approach. (Information Systems Research, 1995)
Authors: Abstract:
    Software development projects are typically team efforts, wherein groups of specialists work toward the common goal of building a software system. The individual efforts of team members need to be coordinated to ensure product quality and effectiveness of the team. In this paper we model the process of coordination in the construction phase of incrementally developed, modular software systems. The analytical model proposed here supports macro-level decisions regarding the development team size and the coordination policy, based upon micro-level interactions between the modules in a system. The objective in this model is to minimize the effort spent on coordination activities subject to the requirement that the system must be completed within a specified period. Results from the model are used to examine coordination related trade-offs. We show that: (1) more complex systems need a higher level of coordination than simpler ones, (2) if the time available for construction reduces, it is optimal to reduce the level of coordination, and (3) marginal productive output is a diminishing function of team size. The sensitivity of the analytical model with respect to its assumptions is studied by constructing a set of simulation experiments where these assumptions are relaxed. The results of these experiments provide support in establishing the robustness of the analytical model.
Improving the Performance Stability of Inductive Expert Systems Under Input Noise. (Information Systems Research, 1995)
Authors: Abstract:
    Inductive expert systems typically operate with imperfect or noisy input attributes. We study design differences in inductive expert systems arising from implicit versus explicit handling of input noise. Most previous approaches use an implicit approach wherein inductive expert systems are constructed using input data of quality comparable to problems the system will be called upon to solve. We develop an explicit algorithm (ID3<sub>ecp</sub>) that uses a clean (without input errors) training set and an explicit measure of the input noise level and compare it to a traditional implicit algorithm, ID3<sub>p</sub> (the ID3 algorithm with the pessimistic pruning procedure). The novel feature of the explicit algorithm is that it injects noise in a controlled rather than random manner in order to reduce the performance variance due to noise. We show analytically that the implicit algorithm has the same expected partitioning behavior as the explicit algorithm. In contrast, however, the partitioning behavior of the explicit algorithm is shown to be more stable (i.e., lower variance) than the implicit algorithm. To extend the analysis to the predictive performance of the algorithms, a set of simulation experiments is described in which the average performance and coefficient of variation of performance of both algorithms are studied on real and artificial data sets. The experimental results confirm the analytical results and demonstrate substantial differences in stability of performance between the algorithms especially as the noise level increases.
Inductive Expert System Design: Maximizing System Value. (Information Systems Research, 1993)
Authors: Abstract:
    There is a growing interest in the use of induction to develop a special class of expert systems known as inductive expert systems. Existing approaches to develop inductive expert systems do not attempt to maximize system value and may therefore be of limited use to firms. We present an induction algorithm that seeks to develop inductive expert systems that maximize value. The task of developing an inductive expert system is looked upon as one of developing an optimal sequential information acquisition strategy. Information is acquired to reduce uncertainty only if the benefits gained from acquiring the information exceed its cost. Existing approaches ignore the costs and benefits of acquiring information. We compare the systems developed by our algorithm with those developed by the popular 1D3 algorithm, in addition. we present results from an extensive set of experiments that indicate that our algorithm will result in more valuable systems than the 1D3 algorithm and the 1D3 algorithm with pessimistic pruning.