Recent years have seen an increasing attention to social aspects of software engineering, including studies of emotions and sentiments experienced and expressed by the software developers. Most of these studies reuse existing sentiment analysis tools such as SENTISTRENGTH and NLTK. However, these tools have been trained on product reviews and movie reviews and, therefore, their results might not be applicable in the software engineering domain. In this paper we study whether the sentiment analysis tools agree with the sentiment recognized by human evaluators (as reported in an earlier study) as well as with each other. Furthermore, we evaluate the impact of the choice of a sentiment analysis tool on software engineering studies by conducting a simple study of differences in issue resolution times for positive, negative and neutral texts. We repeat the study for seven datasets (issue trackers and STACK OVERFLOW questions) and different sentiment analysis tools and observe that the disagreement between the tools can lead to diverging conclusions. Finally, we perform two replications of previously published studies and observe that the results of those studies cannot be confirmed when a different sentiment analysis tool is used.

No abstract

The particle bound mercury (PBM) in urban-industrial areas is mainly of anthropogenic origin, and is derived from two principal sources: Hg bound to particulate matter directly emitted by industries and power generation plants, and adsorption of gaseous elemental mercury (GEM) and gaseous oxidized mercury (GOM) on air particulates from gas or aqueous phases. Here, we measured the Hg isotope composition of PBM in PM 10 samples collected from three locations, a traffic junction, a waste incineration site and an industrial site in Kolkata, the largest metropolis in Eastern India. Sampling was carried out in winter and monsoon seasons between 2013-2015. The objective was to understand whether the isotope composition of the PBM represents source composition. The PBM collected from the waste burning site showed little mass independent fractionation (MIF) (D 199 Hg = + 0.12 to -0.11‰), similar to the signature in liquid Hg and Hg ores around the world with no seasonal variations. Samples from the industrial site showed mostly negative MDF and MIF (δ 202 Hg = -1.34 to -3.48 ‰ and D 199 Hg = + 0.01 to -0.31‰). The MDF is consistent with PBM generated by coal combustion however, the MIF is 0.15‰ more negative compared to the Hg isotope ratios in Indian coals. The traffic junction PBM is probably not produced in situ, but has travelled some distances from nearby industrial sources. The longer residence time of this PBM in the atmosphere has resulted in-aerosol aqueous photoreduction. Thus, the MIF displays a larger range (D 199 Hg = + 0.33 to -0.30‰) compared to the signature from the other sites and with more positive values in the humid monsoon season. Different Hg isotopic signature of PBM in the three different sampling locations within the same city indicates that both source and post emission atmospheric transformations play important roles in determining isotopic signature of PBM.

Abstract-Recent years have seen an increasing attention to social aspects of software engineering, including studies of emotions and sentiments experienced and expressed by the software developers. Most of these studies reuse existing sentiment analysis tools such as SentiStrength and NLTK. However, these tools have been trained on product reviews and movie reviews and, therefore, their results might not be applicable in the software engineering domain.In this paper we study whether the sentiment analysis tools agree with the sentiment recognized by human evaluators (as reported in an earlier study) as well as with each other. Furthermore, we evaluate the impact of the choice of a sentiment analysis tool on software engineering studies by conducting a simple study of differences in issue resolution times for positive, negative and neutral texts. We repeat the study for seven datasets (issue trackers and STACK OVERFLOW questions) and different sentiment analysis tools and observe that the disagreement between the tools can lead to contradictory conclusions.

No abstract