Vladimir Kotlyar

Personalization of Supermarket Product Recommendations

We describe a personalized recommender system designed to suggest new products to supermarket shoppers. The recommender functions in a pervasive computing environment, namely, a remote shopping system in which supermarket customers use Personal Digital Assistants (PDAs) to compose and transmit their orders to the store, which assembles them for subsequent pickup. The recommender is meant to provide an alternative source of new ideas for customers who now visit the store less frequently. Recommendations are generated by matching products to customers based on the expected appeal of the product and the previous spending of the customer. Associations mining in the product domain is used to determine relationships among product classes for use in characterizing the appeal of individual products. Clustering in the customer domain is used to identify groups of shoppers with similar spending histories. Cluster-specific lists of popular products are then used as input to the matching process.The recommender is currently being used in a pilot program with several hundred customers. Analysis of results to date have shown a 1.8% boost in program revenue as a result of purchases made directly from the list of recommended products. A substantial fraction of the accepted recommendations are from product classes new to the customer, indicating a degree of willingness to expand beyond present purchase patterns in response to reasonable suggestions.

Solving Alignment Using Elementary Linear Algebra

Data and computation alignment is an important part of compiling sequential programs to architectures with non-uniform memory access times. In this paper, we show that elementary matrix methods can be used to determine communication-free alignment of code and data. We also solve the problem of replicating read-only data to eliminate communication. Our matrix-based approach leads to algorithms which are simpler and faster than existing algorithms for the alignment problem.

Next-generation generic programming and its application to sparse matrix computations

The contributions of this paper are the following.We introduce a new variety of generic programming in which algorithm implementors use a different API than data structure designers, the gap between the APIs being bridged by restructuring compilers. One view of this approach is that it exploits restructuring compiler technology to perform a novel kind of template instantiation. We demonstrate the usefulness of this new generic programming technology by deploying it in a system that generates efficient sparse codes from high-level algorithms and specifications of sparse matrix formats. We argue that sparse matrix formats should be viewed as indexed-sequential access data structures (in the database sense), and show that appropriate abstractions of the index structure of common formats can be conveyed to a restructuring compiler through the type system of a modern language that supports inheritance and templates.

A Relational Approach to the Compilation of Sparse Matrix Programs

We present a relational algebra based framework for compiling efficient sparse matrix code from dense DO-ANY loops and a specification of the representation of the sparse matrix. We present experimental data that demonstrates that the code generated by our compiler achieves performance competitive with that of hand-written codes for important computational kernels.

Compiling Parallel Code for Sparse Matrix Applications

We have developed a framework based on relational algebra for compiling efficient sparse matrix code from dense DO-ANY loops and a specification of the representation of the sparse matrix. In this paper, we show how this framework can be used to generate parallel code, and present experimental data that demonstrates that the code generated by our Bernoulli compiler achieves performance competitive with that of hand-written codes for important computational kernels.

Software implementation of WiMAX on the sandbridge sandblaster platform

This paper describes a Sandbridge Sandblaster system implementation including both hardware and software elements for a WiMAX 802.16e system. The system is implemented on the fully functional multithreaded Sandblaster multiprocessor SB3010 SoC chip. The entire communication protocol, physical layer and MAC, has been implemented in software using pure ANSI C programming language and it executes in real time. In this paper, we also present a radio propagation analysis specific to the Samos island at the workshop location, and the DSP execution performance.

Analog Television, WiMAX and DVB-H on the Same SoC Platform

This paper presents a SW reconfigurable platform, capable of executing in real time both analog and digital television protocols as NTSC, PAL, SECAM and DVB-H. Our platform is also capable of executing WiMAX at 2.9 Mbps in real time, as well as other mobile telephony protocols like 3G UMTS or CDMA200. SW reconfigurability and low power consumption makes our platform suitable for mobile applications. Current analog and digital television systems have been developed in a combination of analog and/or digital hardware due to high computational processing requirements. They are also mostly limited to a single function either analog TV or, digital TV. DSPs in these systems have been limited to speech coding and orchestrating the custom hardware blocks. Despite the fact that in high-performance systems there may be over 2 million logic gates required to implement physical layer processing, the implementation may take many months to finalize. After the logic design is complete, any errors in the design may cause up to a 9 month delay in correcting and refabricating the device. This labor intensive process is counter productive to fast development cycles. In our approach, the entire physical layer is executed in SW using the SB3011 DSP from Sandbridge Technologies (Glossner et al., 2003) allowing fast development cycles and support of multiple functions

A Case Study in Information Delivery to Mass Retail Markets

The growth in the capabilities of hand-held computers coupled with improvements in database systems and communication infrastructure have brought us closer to enabling truly ubiquitous access to information. In this paper, we describe the SmartPad system, a software solution in the consumer space, providing consumers with ubiquitous access to interactive shopping. This solution enables end-users to shop remotely using personalized content, searching and selecting in incremental mode and disconnected from the network. This is achieved by combining in-store and on-line transactions, by leveraging current efforts in data mining, and by delivering information through hand-held devices. SmartPad is currently used by several hundred consumers with very encouraging initial feedback.

Sparse code generation for imperfectly nested loops with dependences

Standard restructuring compiler tools are based on polyhe dral algebra and cannot be used to analyze or restructure sparse matrix codes We have recently shown that tools based on relational algebra can be used to generate an e cient sparse matrix program from the corresponding dense matrix program and a speci cation of the sparse matrix for mat This work was restricted to DO ALL loops and loops with reductions In this paper we extend this approach to loops with dependences Although our results are restricted to Compressed Hyperplane Storage formats they apply to both perfectly nested loops and imperfectly nested loops

Signal Processing Algorithms For Dsp Implementation of Analog TV Receivers

This paper presents a novel power efficient software implementation of the NTSC analog TV receiver using the Sandbridge Technologies software defined radio (SDR) platform. The radio frequency signal specific to a TV channel is down converted to base band and sampled at 14.318 MHz. From this stage, the entire receiver function is executed in SW, including the horizontal and vertical synchronization, and video processing. The analog TV receiver implemented in DSP can coexist with other communication protocols such as WLAN, GSM/GPRS, GPS etc. allowing analog TV broadcasting on mobile platforms