With the population of social multi‐media and website, smart sentiment analysis from sentences or language becomes a new interesting application in the field of smart cities. In this paper, the ...sentiment is automatically analyzed by extracting the keywords from the long‐and‐difficult English sentences. In supervised scenario, we extract the keywords that are more relevant to express emotions according to syntactic relationship and logical structure, and then assign these keywords weights. In semi‐supervised scenario, we combine the key sentence extraction and classifier fusion algorithm to extract key sentences that contain more sentiment words. During extracting key sentences, we consider sentiment word attribute, position attribute, punctuation attribute, and keywords attribute. During classifier fusion stage, we utilize the classifier with highest confidence to decide the final classification result. The experiments, performed on IMDB and Rotten Tomatoes, show that the proposed method performs better than previous ones. Therefore, extracting keywords via syntactic relationship and logical structure is helpful for smart sentiment analysis.
In this paper, teaching evaluation refers to the students' evaluation of teaching. To help complete the teaching evaluation work better, we construct the corpus of teaching evaluation texts and ...complete the sentiment classification on it. The corpus is collected from a university and processed, which includes 10,299 Chinese sentences. The annotators manually label texts according to the rules designed by educational experts. These texts are divided into three categories, which are positive, negative, and neutral. This paper proposes a sentiment classification method for teaching evaluation texts based on Attention and BiLSTM (Bi-directional Long Short-Term Memory) combined with Syntactic Relationships (BLASR). In this model, the syntactic relationships of sentences are fused into the BiLSTM for feature learning. The weights of different words in the sentences are calculated through an attention layer. The sentiment classification of the teaching evaluation texts is completed by a dense layer. The experimental results show that the classification accuracy of BLASR proposed in this paper on the dataset of teaching evaluation texts is 89.04%, which outperforms baselines. It can satisfy the needs of teaching evaluation in colleges.
► Parsing the query to obtain the set of query terms to calculate the Term Proximity (TP) information. ► Applying different TP measures depending on the lexical type of each query term. ► Applying TP ...measures to phrases as well as terms. ► Obtaining consistent results in the worst conditions reported by previous research.
Traditional Information Retrieval (IR) models assume that the index terms of queries and documents are statistically independent of each other, which is intuitively wrong. This paper proposes the incorporation of the lexical and syntactic knowledge generated by a POS-tagger and a syntactic Chunker into traditional IR similarity measures for including this dependency information between terms. Our proposal is based on theories of discourse structure by means of the segmentation of documents and queries into sentences and entities. Therefore, we measure dependencies between entities instead of between terms. Moreover, we handle discourse references for each entity. It has been evaluated on Spanish and English corpora as well as on Question Answering tasks obtaining significant increases.
The treatise On Adverbs features as one of the Scripta Minora which Apollonius Dyscolus devotes to various parts of speech (mérē toû lógou), and is organically related to his Syntax. The Alexandrine ...grammarian strives to define the adverb as an independent part of speech, not only because of its morphology (not being subject to inflexion), but also because of a set of semanto-syntactical features on which its relationship with the other parts of speech is based. The definition of the adverb, given at the very beginning of the treatise, and minutely discussed in what follows, seems to be the result of his study rather than his starting point. This definition is used as the very criterion for the caseanalyses following in the remainder of the treatise, most particularly the long section devoted to the problems of merism. To Apollonius, invariability is not sufficient to consider a word as an adverb; its syntactic behaviour must conform to the initial definition too: •the adverb is a non-inflected word which restrictively or non-restrictively predicates the inflexions in the verbs, without which it cannot make a meaning complete”. This paper aims to throw some light on the main aspects of this definition: the status of the adverb as a non-essential part of speech; its non-inflected form; its predicative function, and the object of this predication, namely the inflexions of the verb; the combinatory restrictions which limit its use; and finally its appropriate position, which should be before the verb. Apollonius’ view is related to a tradition according to which the adverb is as closely connected with the verb as the adjective is with the noun: only its being employed with a verb is taken into account, and not, for instance, with an adjective. However, the grammarian’s originality lies in his rational approach: he always tries to describe linguistic data in accordance with a lógos, rather than with the appearances of common usage, and to define categories in a comprehensive rather than an extensive manner. This reflects his conception of language as something structured and therefore intelligible.
plutôt qu’extensive. Cette attitude reflète sa conception de la langue comme un ensemble structuré, donc intelligible.