Konstantin Bauman

Aspect Based Recommendations: Recommending Items with the Most Valuable Aspects Based on User Reviews

In this paper, we propose a recommendation technique that not only can recommend items of interest to the user as traditional recommendation systems do but also specific aspects of consumption of the items to further enhance the user experience with those items. For example, it can recommend the user to go to a specific restaurant (item) and also order some specific foods there, e.g., seafood (an aspect of consumption). Our method is called Sentiment Utility Logistic Model (SULM). As its name suggests, SULM uses sentiment analysis of user reviews. It first predicts the sentiment that the user may have about the item based on what he/she might express about the aspects of the item and then identifies the most valuable aspects of the users potential experience with that item. Furthermore, the method can recommend items together with those most important aspects over which the user has control and can potentially select them, such as the time to go to a restaurant, e.g. lunch vs. dinner, and what to order there, e.g., seafood. We tested the proposed method on three applications (restaurant, hotel, and beauty & spa) and experimentally showed that those users who followed our recommendations of the most valuable aspects while consuming the items, had better experiences, as defined by the overall rating.

Using Social Sensors for Detecting Emergency Events: A Case of Power Outages in the Electrical Utility Industry

This article presents a novel approach to detecting emergency events, such as power outages, that utilizes social media users as “social sensors” for virtual detection of such events. The proposed new method is based on the analysis of the Twitter data that leads to the detection of Twitter discussions about these emergency events. The method described in the article was implemented and deployed by one of the vendors in the context of detecting power outages as a part of their comprehensive social engagement platform. It was also field tested on Twitter users in an industrial setting and performed well during these tests.

Optimization of click-through rate prediction in the Yandex search engine

The problem of the estimation of the click-through rate on advertisements that are placed on a search-engine results page is discussed. The proposed methods improved the prediction quality (both in terms of likelihood metrics and the principle parameters of the engine). The cases of advertisement displays are considered when the history of an ad is rather short (i.e., advertisements that are considered to be new). The proposed prediction formula takes the dispersion and high risk of displaying a new advertisement into account.

Minimal Peano curve

A Peano curve p(x) with maximum square-to-linear ratio |p(x)−p(y)|2/|x−y| equal to 5 2/3 is constructed; this ratio is smaller than that of the classical Peano-Hilbert curve, whose maximum square-to-linear ratio is 6. The curve constructed is of fractal genus 9 (i.e., it is decomposed into nine fragments that are similar to the whole curve) and of diagonal type (i.e., it intersects a square starting from one corner and ending at the opposite corner). It is proved that this curve is a unique (up to isometry) regular diagonal Peano curve of fractal genus 9 whose maximum square-to-linear ratio is less than 6. A theory is developed that allows one to find the maximum square-to-linear ratio of a regular Peano curve on the basis of computer calculations.