Building explainable systems is a critical problem in the field of Natural Language Processing (NLP), since most machine learning models provide no explanations for the predictions. Existing approaches for explainable machine learning systems tend to focus on interpreting the outputs or the connections between inputs and outputs. However, the fine-grained information (e.g. textual explanations for the labels) is often ignored, and the systems do not explicitly generate the human-readable explanations. To solve this problem, we propose a novel generative explanation framework that learns to make classification decisions and generate fine-grained explanations at the same time. More specifically, we introduce the explainable factor and the minimum risk training approach that learn to generate more reasonable explanations. We construct two new datasets that contain summaries, rating scores, and fine-grained reasons. We conduct experiments on both datasets, comparing with several strong neural network baseline systems. Experimental results show that our method surpasses all baselines on both datasets, and is able to generate concise explanations at the same time.
Product and Overall Rating ExplanationsMonitor, 3.0Positive Generated: very affordable. unique and ergonomic design. good port selection. Positive Golden: unique design. dual hdmi ports. good color quality. energy efficient.Negative Generated: relatively faint on some features. relatively high contrast ratio. no auto port. Negative Golden: expensive. weak light grayscale performance. features are scarce.Neutral Generated: the samsung series is a unique touch-screen monitor featuring a unique design and a nice capacitive picture, but its color and grayscale performance could be better. Neutral Golden: the samsung series is a stylish 27-inch monitor offering good color reproduction and sharp image quality. however, it 's more expensive than most tn monitors and has a limited feature set.
PCR detection of H. pylori in biological specimens is rendered difficult by the extensive polymorphism of H. pylori genes and the suppressed expression of some genes in many strains. The goal of the present study was to (1) define a domain of the 16S rRNA sequence that is both highly conserved among H. pylori strains and also specific to the species, and (2) to develop and validate specific and sensitive molecular methods for the detection of H. pylori. We used a combination of in silico and molecular approaches to achieve sensitive and specific detection of H. pylori in biologic media. We sequenced two isolates from patients living in different continents and demonstrated that a 546-bp domain of the H. pylori 16S rRNA sequence was conserved in those strains and in published sequences. Within this conserved sequence, we defined a 229-bp domain that is 100% homologous in most H. pylori strains available in GenBank and also is specific for H. pylori. This sub-domain was then used to design (1) a set of high quality RT-PCR primers and probe that encompassed a 76-bp sequence and included at least two mismatches with other Helicobacter sp. 16S rRNA; and (2) in situ hybridization antisense probes. The sensitivity and specificity of the approaches were then demonstrated by using gastric biopsy specimens from patients and rhesus monkeys. This H. pylori-specific region of the 16S rRNA sequence is highly conserved among most H. pylori strains and allows specific detection, identification, and quantification of this bacterium in biological specimens.
A general approach is presented for developing small molecule-based fluorogenic probes suitable for no-wash imaging of endogenous kinases in live cells. Probe 1, including a fluorophore-quencher system, was only "turned on" upon reacting with its target kinase Btk, and disclosed Btk's cellular location in live cells without any washing.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.