Abstract: Disclosed is a method for maintaining compact descriptions of the behavior of customers in a database of information relating to customer transactions. Of all possible variables that might be used to characterize customer transactions, a limited number are selected on the basis of discriminating power to define customer profiles. Customers are classified according to distinguishing features of their profiles. A new customer can be initially assigned to a particular class based on data characterizing his first few transactions.

Abstract: A method is provided for determining a bandwidth allocation needed to provide a specified QOS requirement that takes appropriate account of statistical variations for packet streams in a transmission link. In particular, a statistical model of the packet stream is formed using fractional sum difference statistical models and the model is evaluated in respect to synthetically generated traffic streams. The bandwidth allocation approach is specified in terms of the bandwidth, ?, required for a traffic load, ?, subject to the requirements of a maximum queuing delay, ?, and a packet loss limitation parameter, ?. Accordingly, that bandwidth allocation approach is implemented as a statistical model for ? as a function of ?, ? and ?.

Abstract: Disclosed is a method for maintaining compact descriptions of the behavior of customers in a database of information relating to customer transactions. Of all possible variables that might be used to characterize customer transactions, a limited number are selected on the basis of discriminating power to define customer profiles. Customers are classified according to distinguishing features of their profiles. A new customer can be initially assigned to a particular class based on data characterizing his first few transactions.

Abstract: A monitoring technique is provided for diagnosing delays in file transfers between a server and a client via a network, which monitoring technique operates to make measurements of ordinary network traffic on a link unique to a given client/server pair connection. In particular, the invention operates to apply a set of rules to partition the file transfer time into increments attributable to specific identified causes. To that end, the invention intercepts and sorts traffic traversing a predetermined monitor point into a set of flows, wherein each flow consists of all of the traffic between one client and one server. For each flow the packet headers and timestamps of all packets passing in each direction are read. The partitioning of the file transfer times is based on the resulting data.

Abstract: A method is provided for maintaining compact descriptions of the behavior of customers in a database of information relating to customer transactions. Of all possible variables that might be used to characterize customer transactions, a limited number are selected on the basis of discriminating power to define customer profiles. Customers are classified according to distinguishing features of their profiles. A new customer can be initially assigned to a particular class based on data characterizing his first few transactions.

Abstract: A method is disclosed for processing a silicon workpiece including a hybrid thermometer system for measuring and controlling the processing temperature where fabrication materials have been or are being applied to the workpiece. The hybrid thermometer system uses optical reflectance and another thermometer technique, such as a thermocouple and/or a pyrometer. Real-time spectral data are compared to values in a spectrum library to determine the “surface conditions”. A decision is then made based on the surface conditions as to how the temperature is measured, e.g., with optical reflectance, a pyrometer, or a thermocouple, and the temperature is measured using the appropriately selected technique. Utilizing the hybrid thermometer system, the temperature of a silicon workpiece may be accurately measured at low temperatures while accounting for the presence of fabrication materials.

Abstract: A method is provided for determining a bandwidth allocation needed to provide a specified QOS requirement that takes appropriate account of statistical variations for packet streams in a transmission link. In particular, a statistical model of the packet stream is formed using fractional sum difference statistical models and the model is evaluated in respect to synthetically generated traffic streams. The bandwidth allocation approach is specified in terms of the bandwidth, &bgr;, required for a traffic load, &tgr;, subject to the requirements of a maximum queuing delay, &dgr;, and a packet loss limitation parameter, &ohgr;. Accordingly, that bandwidth allocation approach is implemented as a statistical model for &bgr; as a function of &tgr;, &dgr; and &ohgr;.

Abstract: A monitoring technique is provided for diagnosing delays in file transfers between a server and a client via a network, which monitoring technique operates to make measurements of ordinary network traffic on a link unique to a given client/server pair connection. In particular, the invention operates to apply a set of rules to partition the file transfer time into increments attributable to specific identified causes. To that end, the invention intercepts and sorts traffic traversing a predetermined monitor point into a set of flows, wherein each flow consists of all of the traffic between one client and one server. For each flow the packet headers and timestamps of all packets passing in each direction are read. The partitioning of the file transfer times is based on the resulting data.

Abstract: A customer signature or other type of record in a database system is updated using an event-driven estimator based on a model of time-dependent behavior. In a preferred embodiment, a current version of a record affected by a given transaction is retrieved from a memory of the system and an event-driven estimator of at least one of a transaction rate and a period probability for the record is determined based at least in part on a dynamic Poisson timing model having a number of periods and corresponding period-based transaction rates associated therewith. The event-driven estimator may be configured so as to generate an estimated transaction rate {circumflex over (&lgr;)}j,n for period j and a given transaction n, and then to generate an estimated period probability {circumflex over (&pgr;)}j,n for period j as {circumflex over (&lgr;)}j,n/&Sgr;k{circumflex over (&lgr;)}k,n. An updated version of the record may then be generated based on the event-driven estimator.

Abstract: A customer signature or other type of record in a database system is updated using an event-driven estimator based on a model of time-dependent behavior. In a preferred embodiment, a current version of a record affected by a given transaction is retrieved from a memory of the system and an event-driven estimator of at least one of a transaction rate and a period probability for the record is determined based at least in part on a dynamic Poisson timing model having a number of periods and corresponding period-based transaction rates associated therewith. The event-driven estimator may be configured so as to generate an estimated transaction rate {circumflex over (&lgr;)}j,n for period j and a given transaction n, and then to generate an estimated period probability {circumflex over (&pgr;)}j,n for period j as {circumflex over (&lgr;)}j,n/&Sgr;k{circumflex over (&lgr;)}k,n. An updated version of the record may then be generated based on the event-driven estimator.

Abstract: A system and method for optically identifying a product from a reference library of known products based on a reflected spectrum of the product. A broad wavelength light source illuminates the product and a spectrometer receives and forms a plurality of finely spaced wavelengths from the reflected spectrum. A detector optically processes the wavelengths to generate signals proportional to an amount of light received at each of the wavelengths. The signals are normalized and pre-processed to form data sets which relates each of the signals to each of the finely spaced wavelengths. This is performed for all of the different products and compiled. A set of basis functions is then generated for all of the different products and a corresponding set of basis coefficients is generated for each of the different products. This information, along with an electronic label for each product, is stored to form the reference library.