Nielsen Consumer LLC

Methods, systems, articles of manufacture and apparatus to decompose bias are disclosed. A disclosed example non-transitory machine-readable medium includes machine-readable instructions to cause at least one processor circuit to at least decompose at least one bias of first consumer panel services (CPS) data, and adjust a de-noising model based on the at least one bias, the de-noising model to generate retail measurement services (RMS) data based on second CPS data corresponding to a non-cooperating retailer.

Methods, apparatus, systems, and articles of manufacture are disclosed to tag segments in a document. An example apparatus includes processor circuitry to execute machine readable instructions to generate node embeddings for nodes of a graph, the node embeddings based on features extracted from text segments detected in a document, the text segments to be represented by the nodes of the graph; sample edges corresponding to the nodes to generate the graph; generate first updated node embeddings by passing the node embeddings and the graph through layers of a graph neural network, the first updated embeddings corresponding to the node embeddings augmented with neighbor information; generate second updated node embeddings by passing the first updated embeddings through layers of a recurrent neural network, the second updated embeddings corresponding to the first updated node embeddings augmented with sequential information; and classify the text segments based on the second updated node embeddings.

Systems, apparatus, articles of manufacture, and methods are disclosed that recognize products using visual and textual information. An example apparatus includes classification circuitry to classify object facings detected in a shelf image to determine object candidates corresponding to a first object, ones of the object candidates to include a respective image-based metric; cluster circuitry to assign words extracted from the shelf image to the first object; rank adjuster circuitry to generate text-based metrics for the object candidates by comparing respective object characteristics of the object candidates to the words, ones of the text-based metrics corresponding to respective ones of the object candidates; and prediction circuitry to determine final metrics for the object candidates by selecting, for the ones of the object candidates, (a) a respective image-based metric or (b) a combination of the respective image-based metric and a respective text-based metric; and rank the object candidates using the final metrics.

Systems, apparatus, articles of manufacture, and methods are disclosed to detect promotion events. An apparatus includes interface circuitry to obtain aligned sales data corresponding to products, the products corresponding to a market and a category; computer readable instructions; and programmable circuitry to instantiate event detection circuitry to: identify product-level promotions based on the aligned sales data and corresponding baseline data for the products; and group ones of the product-level promotions to identify promotion events; and expansion circuitry to determine uplift factors for different promotion characteristics based on the market and the category; and apply the uplift factors to the product-level promotions and the promotion events to identify incremental sales.

Systems and apparatus to train machine learning models for item coding are disclosed. An example apparatus to pretrain a machine learning model for item coding includes machine-readable instructions; and at least one programmable circuit to at least one of instantiate or execute the machine-readable instructions to generate a data structure based on a product description, the data structure defining a hierarchy of product characteristics; train a machine learning (ML) model based on the data structure and using causal training; and generate, based on the causal training, a pre-trained ML model to output predicted product characteristics for item coding.

Methods, apparatus, systems, and articles of manufacture are disclosed to improve tagging accuracy. An example apparatus includes at least one memory, machine readable instructions, and processor circuitry to execute the machine readable instructions to at least search a first row of a document to identify a first row that includes a first type of entity, search the first row of the document to identify a second type of entity that is missing, search the first row of the document to identify a first integer value, and associate the first row with a product corresponding to the first integer value.

Systems, apparatus, articles of manufacture, and methods are disclosed to label data. An example apparatus includes interface circuitry, machine-readable instructions, and at least one processor circuit to be programmed by the machine-readable instructions to separate label data in a first data set from portions of an image, generate candidate labeled data based on associated ones of unlabeled portions of the image and optical character recognition (OCR) data, generate key performance indicator (KPI) metric values based on a comparison between the candidate labeled data and a second data set, and adjust weights of a model based on the KPI metric values.

Systems, apparatus, articles of manufacture, and methods are disclosed to train models with a dynamic margin. An example apparatus includes interface circuitry, machine-readable instructions, and at least one processor circuit to be programmed by the machine-readable instructions to determine a positive label similarity value and a negative label similarity value, the positive and negative similarity values based on a query embedding, determine a negative-to-positive difference value and a positive-to-negative difference value associated with the positive label similarity value and the negative label similarity value, determine a positive-to-negative difference value associated with the positive label similarity value and the negative label similarity value, and cause training of a label classifier with a loss function having a dynamic margin when the positive-to-negative difference value satisfies a threshold.

Methods, systems, articles of manufacture, and apparatus to recalibrate confidences for image classification are disclosed. An example apparatus to classify an image includes an image crop detector to detect a first image crop from the image, the first image crop corresponding to a first object, a grouping controller to select a second image crop corresponding to a second object at a location of the first object, a prediction generator to, in response to executing a trained model, determine a label corresponding to the first object and a confidence level associated with the label, and a confidence recalibrator to recalibrate the confidence level based on a probability of the first object having a first attribute based on the second object having a second attribute, the confidence level recalibrated to increase an accuracy of the image classification.

Methods, apparatus, systems, and articles of manufacture are disclosed to tag segments in a document. An example apparatus includes processor circuitry to execute machine readable instructions to generate node embeddings for nodes of a graph, the node embeddings based on features extracted from text segments detected in a document, the text segments to be represented by the nodes of the graph; sample edges corresponding to the nodes to generate the graph; generate first updated node embeddings by passing the node embeddings and the graph through layers of a graph neural network, the first updated embeddings corresponding to the node embeddings augmented with neighbor information; generate second updated node embeddings by passing the first updated embeddings through layers of a recurrent neural network, the second updated embeddings corresponding to the first updated node embeddings augmented with sequential information; and classify the text segments based on the second updated node embeddings.