Data mining can discover information hidden within valuable data assets. Knowledge discovery, using advanced information technologies, can uncover veins of surprising, golden insights in a mountain of factual data. Data mining consists of a panoply of powerful tools which are intuitive, easy to explain, understandable, and simple to use. These advanced information technologies include artificial intelligence methods (e.g. expert systems, fuzzy logic, etc.), decision trees, rule induction methods, genetic algorithms and genetic programming, neural networks (e.g. backpropagation, associate memories, etc.), and clustering techniques. The synergy created between data warehousing and data mining allows knowledge seekers to leverage their massive data assets, thus improving the quality and effectiveness of their decisions. The growing requirements for data mining and real time analysis of information will be a driving force in the development of new data warehouse architectures and methods and, conversely, the development of new data mining methods and applications.

No abstract

Unlike other computer-based information systems, expert systems (ES) are characterized by the satisficing and conservative behavior of their users. Shows that the learning curve may be used to model user dependency on ES technology. Even though user dependency relates to ES quality control parameters (for example, Raggad's 13 ES quality attributes) only dynamic or late binding features really affect ES dependency: ES learning capability and ES recommendation anticipation. There is hence a learning race between the system and the user. If user learning prevails, then there will be user defection. If system learning prevails, then there will be system perfection. Proposes failure analysis based on user defection due to the absence or underutilization of machine learning. ES owners can adopt this model to design a subsystem capable of transforming user defection analysis into a strategic plan for ES management.

e18366 Background: For unresectable, well-differentiated, advanced GEP-NET patients somatostatin analogs octreotide LAR (OCT) and lanreotide depot (LAN) are recommended as first-line systemic therapy. Recent evidence suggests that GEP-NET patients may benefit from LAN after OCT prior to modifying treatment class. The goal of this analysis is to quantify treatment escalation costs among patients who progress on OCT. Methods: A decision-tree model was used to assess the cost of treating GEP-NET patients over a 6-month time horizon after they progress on OCT. Drug acquisition (Wholesale Acquisition Costs), administration, grade 3/4 adverse event (AE) management, and patient monitoring costs were considered. Patients could utilize either OCT escalation (30 mg every 3 weeks or 40 or 60 mg every 4 weeks), OCT + peptide receptor radionuclide therapy (PRRT), OCT + liver directed therapy (e.g. transarterial chemoembolization (TACE)), everolimus, or LAN 120mg every 4 weeks. Administration and monitoring test costs were based on CMS physician and laboratory fee schedules, respectively. AE management costs were determined from the CMS Inpatient prospective payment system based on reimbursements corresponding to the associated DRG codes. Results: LAN ($44,462) was found to be cost-saving versus alternatives when used post-OCT. OCT escalations to 30mg every 3 weeks, 40mg every 4 weeks, and its use with liver directed therapy were 2nd, 3rd, and 4th lowest cost treatment options ($52,873, $53,041, and $72,152, respectively); all three had higher drug acquisition and AE management costs. Drug acquisition costs for everolimus were more than twice those of LAN and for PRRT + OCT nearly 4 times that amount. GA-68 Dotatate PET/CT imaging greatly increased the monitoring costs of PRRT + OCT, which were $3,682 higher than the nearest alternative. Conclusions: LAN was the lowest cost treatment option over a 6-month interval for GEP-NET patients who progressed on OCT, however clinical appropriateness must be considered when transitioning patients. Evidence for the appropriate clinical strategy of sequencing of patients is still needed.

No abstract