Abstract: Sentiment analysis or opinion mining is one of the major tasks of NLP (Natural Language Processing). Sentiment analysis has gain much attention in recent years. In this paper, we aim to tackle the problem of sentiment polarity categorization, which is one of the fundamental problems of sentiment analysis. A general process for sentiment polarity categorization is proposed with detailed process descriptions. Data used in this study are online product reviews collected from Amazon.com. Experiments for both sentence-level categorization and review-level categorization are performed with promising outcomes. At last, we also give insight into our future work on sentiment analysis.

Abstract: Human performance at categorizing natural visual images surpasses automatic algorithms, but how and when this function arises and develops remain unanswered. We recorded scalp electrical brain activity in 4-6 months infants viewing images of objects in their natural background at a rapid rate of 6 images/second (6 Hz). Widely variable face images appearing every 5 stimuli generate an electrophysiological response over the right hemisphere exactly at 1.2 Hz (6 Hz/5). This face-selective response is absent for phase-scrambled images and therefore not due to low-level information. These findings indicate that right lateralized face-selective processes emerge well before reading acquisition in the infant brain, which can perform figure-ground segregation and generalize face-selective responses across changes in size, viewpoint, illumination as well as expression, age and gender. These observations made with a highly sensitive and objective approach open an avenue for clarifying the developmental course of natural image categorization in the human brain.

Abstract: We propose a novel data augmentation approach to enhance computational behavioral analysis using social media text. In particular, we collect a Twitter corpus of the descriptions of annoying behaviors using the #petpeeve hashtags. In the qualitative analysis, we study the language use in these tweets, with a special focus on the fine-grained categories and the geographic variation of the language. In quantitative analysis, we show that lexical and syntactic features are useful for automatic categorization of annoying behaviors, and frame-semantic features further boost the performance; that leveraging large lexical embeddings to create additional training instances significantly improves the lexical model; and incorporating frame-semantic embedding achieves the best overall performance.

Abstract: Categorization plays a fundamental role in organizing daily interactions with the social world. However, there is increasing recognition that social categorization is often complex, both because category membership can be ambiguous (e.g., multiracial or transgender identities) and because different categorical identities (e.g., race and gender) may interact to determine the meaning of category membership. These complex identities simultaneously impact social perceivers' impressions and social targets' own experiences of identity, thereby shaping perceptions, experiences, and interactions in fundamental ways. This review examines recent research on the perception and experience of the complex, multifaceted identities that both complicate and enrich our lives. Although research has historically tended to focus more on difficulties and challenges associated with multiple identities, increasing attention is being paid to opportunities that emerge from the possession of identities that include multiple distinc...

Abstract: Fine-grained categorization, which aims to distinguish subordinate-level categories such as bird species or dog breeds, is an extremely challenging task. This is due to two main issues: how to localize discriminative regions for recognition and how to learn sophisticated features for representation. Neither of them is easy to handle if there is insufficient labeled data. We leverage the fact that a subordinate-level object already has other labels in its ontology tree. These "free" labels can be used to train a series of CNN-based classifiers, each specialized at one grain level. The internal representations of these networks have different region of interests, allowing the construction of multi-grained descriptors that encode informative and discriminative features covering all the grain levels. Our multiple granularity framework can be learned with the weakest supervision, requiring only image-level label and avoiding the use of labor-intensive bounding box or part annotations. Experimental results on three challenging fine-grained image datasets demonstrate that our approach outperforms state-of-the-art algorithms, including those requiring strong labels.

Abstract: Categorization with basic color terms is an intuitive and universal aspect of color perception. Yet research on visual working memory capacity has largely assumed that only continuous estimates within color space are relevant to memory. As a result, the influence of color categories on working memory remains unknown. We propose a dual content model of color representation in which color matches to objects that are either present (perception) or absent (memory) integrate category representations along with estimates of specific values on a continuous scale (“particulars”). We develop and test the model through 4 experiments. In a first experiment pair, participants reproduce a color target, both with and without a delay, using a recently influential estimation paradigm. In a second experiment pair, we use standard methods in color perception to identify boundary and focal colors in the stimulus set. The main results are that responses drawn from working memory are significantly biased away from category boundaries and toward category centers. Importantly, the same pattern of results is present without a memory delay. The proposed dual content model parsimoniously explains these results, and it should replace prevailing single content models in studies of visual working memory. More broadly, the model and the results demonstrate how the main consequence of visual working memory maintenance is the amplification of category related biases and stimulus-specific variability that originate in perception.

Abstract: Fine-grained visual categorization (FGVC) is to categorize objects into subordinate classes instead of basic classes. One major challenge in FGVC is the co-occurrence of two issues: 1) many subordinate classes are highly correlated and are difficult to distinguish, and 2) there exists the large intra-class variation (e.g., due to object pose). This paper proposes to explicitly address the above two issues via distance metric learning (DML). DML addresses the first issue by learning an embedding so that data points from the same class will be pulled together while those from different classes should be pushed apart from each other; and it addresses the second issue by allowing the flexibility that only a portion of the neighbors (not all data points) from the same class need to be pulled together. However, feature representation of an image is often high dimensional, and DML is known to have difficulty in dealing with high dimensional feature vectors since it would require O(d2) for storage and O(d3) for optimization. To this end, we proposed a multi-stage metric learning framework that divides the large-scale high dimensional learning problem to a series of simple subproblems, achieving O(d) computational complexity. The empirical study with FVGC benchmark datasets verifies that our method is both effective and efficient compared to the state-of-the-art FGVC approaches.

Abstract: In this paper, we consider the task of text categorization as a graph classification problem. By representing textual documents as graph-of-words instead of historical n-gram bag-of-words, we extract more discriminative features that correspond to long-distance n-grams through frequent subgraph mining. Moreover, by capitalizing on the concept of k-core, we reduce the graph representation to its densest part – its main core – speeding up the feature extraction step for little to no cost in prediction performances. Experiments on four standard text classification datasets show statistically significant higher accuracy and macro-averaged F1-score compared to baseline approaches.

Abstract: People make sense of objects and events around them by classifying them into identifiable categories. The extent to which language affects this process has been the focus of a long-standing debate: Do different languages cause their speakers to behave differently? Here, we show that fluent German-English bilinguals categorize motion events according to the grammatical constraints of the language in which they operate. First, as predicted from cross-linguistic differences in motion encoding, bilingual participants functioning in a German testing context prefer to match events on the basis of motion completion to a greater extent than do bilingual participants in an English context. Second, when bilingual participants experience verbal interference in English, their categorization behavior is congruent with that predicted for German; when bilingual participants experience verbal interference in German, their categorization becomes congruent with that predicted for English. These findings show that language effects on cognition are context-bound and transient, revealing unprecedented levels of malleability in human cognition.

Abstract: Existing fine-grained visual categorization methods often suffer from three challenges: lack of training data, large number of fine-grained categories, and high intraclass vs. low inter-class variance. In this work we propose a generic iterative framework for fine-grained categorization and dataset bootstrapping that handles these three challenges. Using deep metric learning with humans in the loop, we learn a low dimensional feature embedding with anchor points on manifolds for each category. These anchor points capture intra-class variances and remain discriminative between classes. In each round, images with high confidence scores from our model are sent to humans for labeling. By comparing with exemplar images, labelers mark each candidate image as either a "true positive" or a "false positive". True positives are added into our current dataset and false positives are regarded as "hard negatives" for our metric learning model. Then the model is retrained with an expanded dataset and hard negatives for the next round. To demonstrate the effectiveness of the proposed framework, we bootstrap a fine-grained flower dataset with 620 categories from Instagram images. The proposed deep metric learning scheme is evaluated on both our dataset and the CUB-200-2001 Birds dataset. Experimental evaluations show significant performance gain using dataset bootstrapping and demonstrate state-of-the-art results achieved by the proposed deep metric learning methods.

Abstract: In this paper, a new fine-grained image categorization system that improves spatial pyramid matching is developed. In this method, multiple cells are combined into cellets in the proposed categorization model, which are highly responsive to an object's fine categories. The object components are represented by cellets that can connect spatially adjacent cells within the same pyramid level. Here, image categorization can be formulated as the matching between the cellets of corresponding images. Toward an effective matching process, an active learning algorithm that can effectively select a few representative cells for constructing the cellets is adopted. A linear-discriminant-analysis-like scheme is employed to select discriminative cellets. Then, fine-grained image categorization can be conducted with a trained linear support vector machine. Experimental results on three real-world data sets demonstrate that our proposed system outperforms the state of the art.

Abstract: This paper presents experiments extending the work of Ba et al. (2014) on recurrent neural models for attention into less constrained visual environments, specifically fine-grained categorization on the Stanford Dogs data set. In this work we use an RNN of the same structure but substitute a more powerful visual network and perform large-scale pre-training of the visual network outside of the attention RNN. Most work in attention models to date focuses on tasks with toy or more constrained visual environments, whereas we present results for fine-grained categorization better than the state-of-the-art GoogLeNet classification model. We show that our model learns to direct high resolution attention to the most discriminative regions without any spatial supervision such as bounding boxes, and it is able to discriminate fine-grained dog breeds moderately well even when given only an initial low-resolution context image and narrow, inexpensive glimpses at faces and fur patterns. This and similar attention models have the major advantage of being trained end-to-end, as opposed to other current detection and recognition pipelines with hand-engineered components where information is lost. While our model is state-of-the-art, further work is needed to fully leverage the sequential input.

Abstract: Recognizing an object takes just a fraction of a second, less than the blink of an eye. Applying multivariate pattern analysis, or “brain decoding”, methods to magnetoencephalography (MEG) data has allowed researchers to characterize, in high temporal resolution, the emerging representation of object categories that underlie our capacity for rapid recognition. Shortly after stimulus onset, object exemplars cluster by category in a high-dimensional activation space in the brain. In this emerging activation space, the decodability of exemplar category varies over time, reflecting the brain’s transformation of visual inputs into coherent category representations. How do these emerging representations relate to categorization behavior? Recently it has been proposed that the distance of an exemplar representation from a categorical boundary in an activation space is critical for perceptual decision-making, and that reaction times should therefore correlate with distance from the boundary. The predictions of this distance hypothesis have been born out in human inferior temporal cortex (IT), an area of the brain crucial for the representation of object categories. When viewed in the context of a time varying neural signal, the optimal time to “read out” category information is when category representations in the brain are most decodable. Here, we show that the distance from a decision boundary through activation space, as measured using MEG decoding methods, correlates with reaction times for visual categorization during the period of peak decodability. Our results suggest that the brain begins to read out information about exemplar category at the optimal time for use in choice behaviour, and support the hypothesis that the structure of the representation for objects in the visual system is partially constitutive of the decision process in recognition.

Abstract: Categorizations of multiracial individuals provide insight into the psychological mechanisms driving social stratification, but few studies have explored the interplay of cognitive and motivational underpinnings of these categorizations In the present study, we integrated research on racial essentialism (ie, the belief that race demarcates unobservable and immutable properties) and negativity bias (ie, the tendency to weigh negative entities more heavily than positive entities) to explain why people might exhibit biases in the categorization of multiracial individuals As theorized, racial essentialism, both dispositional (Study 1) and experimentally induced (Study 2), led to the categorization of Black-White multiracial individuals as Black, but only among individuals evaluating Black people more negatively than White people These findings demonstrate how fundamental cognitive and motivational biases interact to influence the categorization of multiracial individuals

Abstract: Allocating attentional resources to currently relevant information in a dynamically changing environment is critical to goal-directed behavior. Previous studies in nonhuman primates (NHPs) have demonstrated modulation of neural representations of stimuli, in particular visual categorizations, by behavioral significance in the lateral prefrontal cortex. In the human brain, a network of frontal and parietal regions, the "multiple demand" (MD) system, is involved in cognitive and attentional control. To test for the effect of behavioral significance on categorical discrimination in the MD system in humans, we adapted a previously used task in the NHP and used multivoxel pattern analysis for fMRI data. In a cued-detection categorization task, participants detected whether an image from one of two target visual categories was present in a display. Our results revealed that categorical discrimination is modulated by behavioral relevance, as measured by the distributed pattern of response across the MD network. Distinctions between categories with different behavioral status (e.g., a target and a nontarget) were significantly discriminated. Category distinctions that were not behaviorally relevant (e.g., between two targets) were not discriminated. Other aspects of the task that were orthogonal to the behavioral decision did not modulate categorical discrimination. In a high visual region, the lateral occipital complex, modulation by behavioral relevance was evident in its posterior subregion but not in the anterior subregion. The results are consistent with the view of the MD system as involved in top-down attentional and cognitive control by selective coding of task-relevant discriminations. Significance statement: Control of cognitive demands fundamentally involves flexible allocation of attentional resources depending on a current behavioral context. Essential to such a mechanism is the ability to select currently relevant information and at the same time filter out information that is irrelevant. In an fMRI study, we measured distributed patterns of activity for objects from different visual categories while manipulating the behavioral relevance of the categorical distinctions. In a network of frontal and parietal cortical regions, the multiple-demand (MD) network, patterns reflected category distinctions that were relevant to behavior. Patterns could not be used to make task-irrelevant category distinctions. These findings demonstrate the ability of the MD network to implement complex goal-directed behavior by focused attention.

Abstract: Speech perception depends on long-term representations that reflect regularities of the native language. However, listeners rapidly adapt when speech acoustics deviate from these regularities due to talker idiosyncrasies such as foreign accents and dialects. To better understand these dual aspects of speech perception, we probe native English listeners' baseline perceptual weighting of 2 acoustic dimensions (spectral quality and vowel duration) toward vowel categorization and examine how they subsequently adapt to an "artificial accent" that deviates from English norms in the correlation between the 2 dimensions. At baseline, listeners rely relatively more on spectral quality than vowel duration to signal vowel category, but duration nonetheless contributes. Upon encountering an "artificial accent" in which the spectral-duration correlation is perturbed relative to English language norms, listeners rapidly down-weight reliance on duration. Listeners exhibit this type of short-term statistical learning even in the context of nonwords, confirming that lexical information is not necessary to this form of adaptive plasticity in speech perception. Moreover, learning generalizes to both novel lexical contexts and acoustically distinct altered voices. These findings are discussed in the context of a mechanistic proposal for how supervised learning may contribute to this type of adaptive plasticity in speech perception.

Abstract: Humans understand the world by classifying objects into an appropriate level of categories. This process is often automatic and subconscious. Psychologists and linguists call it as Basic-level Categorization (BLC). BLC can benefit lots of applications such as knowledge panel, advertising and recommendation. However, how to quantify basic-level concepts is still an open problem. Recently, much work focuses on constructing knowledge bases or semantic networks from web scale text corpora, which makes it possible for the first time to analyze computational approaches for deriving BLC. In this paper, we introduce a method based on typicality and PMI for BLC. We compare it with a few existing measures such as NPMI and commute time to understand its essence, and conduct extensive experiments to show the effectiveness of our approach. We also give a real application example to show how BLC can help sponsored search.

Abstract: Prior research on the development of race‐based categorization has concluded that children understand the perceptual basis of race categories from as early as age 4 (e.g. Aboud, ). However, such work has rarely separated the influence of skin color from other physiognomic features considered by adults to be diagnostic of race categories. In two studies focusing on Black–White race categorization judgments in children between the ages of 4 and 9, as well as in adults, we find that categorization decisions in early childhood are determined almost entirely by attention to skin color, with attention to other physiognomic features exerting only a small influence on judgments as late as middle childhood. We further find that when skin color cues are largely eliminated from the stimuli, adults readily shift almost entirely to focus on other physiognomic features. However, 6‐ and 8‐year‐old children show only a limited ability to shift attention to facial physiognomy and so perform poorly on the task. These results demonstrate that attention to ‘race’ in younger children is better conceptualized as attention to skin color, inviting a reinterpretation of past work focusing on children's race‐related cognition. What perceptual features do children use to categorize by race? Past research has suggested that adult‐like abilities to racially classify emerge quite early in development. Our findings suggest this is not the case; younger children rely almost entirely on skin color, with little or no attention to other aspects of facial physiognomy.

Abstract: Although individual differences in category-learning tasks have been explored, the observed differences have tended to represent different instantiations of general processes (e.g., learners rely upon different cues to develop a rule) and their consequent representations. Additionally, studies have focused largely on participants’ categorizations of transfer items to determine the representations that they formed. In the present studies, we used a convergent-measures approach to examine participants’ categorizations of transfer items in addition to their self-reported learning orientations and response times on transfer items, and in doing so, we garnered evidence that qualitatively distinct approaches in explicit strategies for category learning (i.e., memorization vs. abstracting an articulable rule) and consequent representations might emerge in a single task. Participants categorized instances that followed a categorization rule (in Study 1, we used a relational rule; in Study 2, an additional task with a single-feature rule). Critically, for both tasks, some transfer items differed from trained instances on only one attribute (but otherwise were perceptually similar), rendering the item a member of the opposing category on the basis of the rule (i.e., termed ambiguous items). Some learners categorized ambiguous items on the basis of perceptual similarity, whereas others categorized them on the basis of an abstracted rule. Self-reported learning orientation (i.e., memorization vs. rule abstraction) predicted categorizations and response times on transfer items. Differences in learning orientations were not associated with performance on other cognitive measures (i.e., working memory capacity and Raven’s Advanced Progressive Matrices). This work suggests that individuals may have different predispositions toward memorization versus rule abstraction in a single categorization task.

Abstract: Purpose – The purpose of this paper is to explore the interrelation between language differences, media choice and social categorization in global virtual teams (GVTs). Design/methodology/approach – An ethnographic field work was conducted in a Finnish multinational corporation (MNC). The study included interviews, observations, and language proficiency assessment of 27 GVT members located in five European countries. Findings – In GVTs, the combination of language proficiency differences and verbal media (e.g. telephone) tends to lead to social categorization, while a similar effect was not found when GVT members chose written media (e.g. e-mail). Research limitations/implications – The qualitative study only consisted of GVTs from one MNC, and thus the empirical findings might not be generalizable to other MNCs. Therefore, quantitative studies that can add to the robustness of the exploratory findings could be a worthwhile endeavour. Practical implications – Language training should be provided to GVT me...

Abstract: Categorization is known to play an important role in organizations because categories ‘frame’ situations in particular ways, informing managerial sensemaking and enabling managerial intervention. In this article, we advance existing work by examining the role of categorization practices in discursive leadership during periods of strategic change. Drawing on data from an ethnographic action research study of a strategic change initiative in a multi-national corporation, we use membership categorization analysis to develop a framework for studying ‘category predicates’ − defined as the stock of organizational knowledge and associated reasoning procedures concerning the kinds of activities, attributes, rights, responsibilities, expectations, and so on, that are ‘tied’ or ‘bound’ to organizational categories. Our analysis shows that discursive leadership enabled a radical shift in sensemaking about organizational structure categories through a process of ‘frame-breaking’ and ‘re-framing’. In so doing, the leader co-constructed a ‘definition of the situation’ that built a compelling vision and concrete plan for strategic change. We go on to trace the organizational consequences and material outcomes of this shift in sensemaking for the company in question. We conclude by arguing that ‘category predication work’ comprises a key leadership competence and plays an important role in organizational and strategic change processes.

Abstract: It has been proposed that much of conceptual knowledge is acquired through situated conceptualization, such that both external (e.g., agents, objects, events) and internal (e.g., emotions, introspections) environments are considered important (Barsalou, 2003). To evaluate this proposal, we characterized two dimensions by which situated conceptualization may be measured and which should have different relevance for abstract and concrete concepts; namely, emotional experience (i.e., the ease with which words evoke emotional experience; Newcombe, Campbell, Siakaluk, & Pexman, 2012) and context availability (i.e., the ease with which words evoke contexts in which their referents may appear; Schwanenflugel & Shoben, 1983). We examined the effects of these two dimensions on abstract and concrete word processing in verbal semantic categorization (VSCT) and naming tasks. In the VSCT, emotional experience facilitated processing of abstract words but inhibited processing of concrete words, whereas context availability facilitated processing of both types of words. In the naming task in which abstract words and concrete words were not blocked by emotional experience, context availability facilitated responding to only the abstract words. In the naming task in which abstract words and concrete words were blocked by emotional experience, emotional experience facilitated responding to only the abstract words, whereas context availability facilitated responding to only the concrete words. These results were observed even with several lexical (e.g., frequency, age of acquisition) and semantic (e.g., concreteness, arousal, valence) variables included in the analyses. As such, the present research suggests that emotional experience and context availability tap into different aspects of situated conceptualization and make unique contributions to the representation and processing of abstract and concrete concepts.

Abstract: We investigated how linguistic intention affects the time course of visual word recognition by comparing the brain's electrophysiological response to a word's lexical frequency, a well-established psycholinguistic marker of lexical access, when participants actively retrieve the meaning of the written input semantic categorization versus a situation where no language processing is necessary ink color categorization. In the semantic task, the ERPs elicited by high-frequency words started to diverge from those elicited by low-frequency words as early as 120 msec after stimulus onset. On the other hand, when categorizing the colored font of the very same words in the color task, word frequency did not modulate ERPs until some 100 msec later 220 msec poststimulus onset and did so for a shorter period and with a smaller scalp distribution. The results demonstrate that, although written words indeed elicit automatic recognition processes in the brain, the speed and quality of lexical processing critically depends on the top-down intention to engage in a linguistic task.

Abstract: Ensemble summary statistics represent multiple objects on the high level of abstraction-that is, without representing individual features and ignoring spatial organization. This makes them especially useful for the rapid visual categorization of multiple objects of different types that are intermixed in space. Rapid categorization implies our ability to judge at one brief glance whether all visible objects represent different types or just variants of one type. A framework presented here states that processes resembling statistical tests can underlie that categorization. At an early stage (primary categorization), when independent ensemble properties are distributed along a single sensory dimension, the shape of that distribution is tested in order to establish whether all features can be represented by a single or multiple peaks. When primary categories are separated, the visual system either reiterates the shape test to recognize subcategories (in-depth processing) or implements mean comparison tests to match several primary categories along a new dimension. Rapid categorization is not free from processing limitations; the role of selective attention in categorization is discussed in light of these limitations.