Victor Soares Bursztyn

Northwestern University | NU · Department of Electrical Engineering and Computer Science

The goal of this work is to analyze two of the most central activities in the life of a congressman: raising funds and voting bills. We investigate the Brazilian Congress to shed light on the relationships between the donations received by congressmen elected in 2014 and their voting behaviors during the year of 2015. We merged publicly available d...

There is growing concern about the use of social platforms to push political narratives during elections. One very recent case is Brazil's, where WhatsApp is now widely perceived as a key enabler of the far-right's rise to power. In this paper, we perform a large-scale analysis of partisan WhatsApp groups to shed light on how both right-wingers and...

Two fundamental activities in the career of a congressman are raising campaign funds and voting on bills. Characterizing the interplay between these two activities is crucial for democracy as it provides explicit feedback to voters on potential conflicts of interest. In this work, we investigate the Brazilian Congress to shed light on the relations...

How to usefully encode compositional task structure has long been a core challenge in AI. Recent work in chain of thought prompting has shown that for very large neural language models (LMs), explicitly demonstrating the inferential steps involved in a target task may improve performance over end-to-end learning that focuses on the target task alon...

Based on recent advances in realistic language modeling (GPT-3) and cross-modal representations (CLIP), Gaudí was developed to help designers search for inspirational images using natural language. In the early stages of the design process, with the goal of eliciting a client's preferred creative direction, designers will typically create thematic...

Conversations aimed at determining good recommendations are iterative in nature. People often express their preferences in terms of a critique of the current recommendation (e.g., "It doesn't look good for a date"), requiring some degree of common sense for a preference to be inferred. In this work, we present a method for transforming a user criti...

Product planning techniques are provided that recommend compositions of product features for weighted heterogeneous consumer segments using regression trees. An exemplary method comprises obtaining historical consumer data comprising product preferences for existing product items for multiple consumer segments ; obtaining product features indicatin...

We have developed a conversational recommendation system designed to help users navigate through a set of limited options to find the best choice. Unlike many internet scale systems that use a singular set of search terms and return a ranked list of options from amongst thousands, our system uses multi-turn user dialog to deeply understand the user...

Content planning techniques are provided that recommend content features based on the investment interest of advertisers in various audience segments and historical audience measurements . An exemplary method comprises obtaining historical data comprising content preferences indicating a performance metric for each pair of a plurality of content it...

One major challenge in large-scale experiments is the analytical capacity to contrast ongoing results with domain knowledge. We approach this challenge by constructing a domain-specific knowledge base, which is queried during workflow execution. We introduce K-Chiron, an integrated solution that combines a state-of-the-art automatic knowledge base...

Finding the best way to make the data well-grounded in domain knowledge is an important challenge in large-scale experiments. While this responsibility inherently depends on the role of domain experts, the large amount of domain-related facts and their associated complexity increasingly makes this role too labor-intensive, susceptible to errors, an...

One major challenge in large-scale experiments is the analytical capacity to contrast ongoing results with domain knowledge. We approach this challenge by constructing a domain-specific knowledge base, which is queried during workflow execution. We introduce K-Chiron, an integrated solution that combines a state-of-the-art automatic knowledge base...

This works presents a methodology to suggest potential relationships among the elements in the scientific collaboration network. The proposed approach takes into account not only the structure of the relationships among the individuals that constitute the network, but also the content of the information flow propagated in it, modeled from the docum...

Figura 1: (a) Imagem renderizada do aerogerador modelado em Blender (b) Réplica do modelo em miniatura (c) Aerogerador original instalado em Mucuripe (Fortaleza – CE) Abstract This paper describes an energy industry related issue, focused in the machinery of wind power. Energy area is specially challenging when it comes to share technical informati...

This work presents, 3DbyStep, an interactive authoring tool for developing 3D training contents. The application is able to combine text, pictures, sound and 3D models in an integrated environment. The produced contents is suitable for using as a standard presentation and also as a means to support self-study in the sense that it can be navigated i...

This paper describes a work that includes the development and implementation of a practical and efficient methodology to construct a knowledge extraction environment that contemplates the search of information from Portuguese language Web sites. The application has much functionality in text mining, such as similarities and differences identificati...