Sentiment Analysis is currently one of the most studied research fields. Its aim is to analyze people‟s sentiments, opinions, attitudes etc., towards different elements such as topics, products, individuals, organizations, and services. Sentiment analysis can be achieved by machine learning or lexical based methodologies or a combination of both. Recent research shows that domain specific lexicons perform better as compared to domain independent lexicons. In an effort to improve the performance of domain independent lexicons, this research incorporates machine learning with a lexical based approach, introducing a new approach called SWIMS, to determine the feature weight based on a well-known general-purpose sentiment lexicon, SentiWordNet. Support vector machine is used to learn the feature weights and an intelligent model selection approach is employed in SWIMS in order to enhance the classification performance. The features are selected based on their subjectivity and the effects of feature selection with respect to their part of speech information are studied extensively. Seven benchmark datasets have been used in this research, including large movie review dataset, multi-domain sentiment dataset and Cornell movie review dataset, all of which are freely available online for research purposes. In-depth performance comparison is conducted with the state of the art machine learning approaches, lexical based methodologies, and other tools used for sentiment detection. The proposed SWIMS approach attained accuracy, precision, recall and f-measure values of 83.06%, 83.30%, 82.83% and 83.03% respectively, averaged over the seven datasets. The evaluation of performance measures proves that the proposed approach outperforms other techniques for sentiment analysis.
اک دفعہ دا ذکر اے کہ چار پاگل بازار وچ ٹرے جا رہے سن۔ اوہناں دی نظر حلوہ کدو اتے پئی۔ اوہناں آکھیا ایہہ گھوڑے دا انڈہ اے۔ جے ایس نوں اسیں اپنے کھیت وچ بیج لئے تاں گھوڑے اُگن گے۔ اوہناں دکان دار کولوں چار وڈے وڈے حلوہ کدو خریدے تے اپنے پنڈ پرت آئے۔
اگلے دن چاروں اپنے کھیت ول جاندے نیں۔ اوہناں بہت محنت نال ہل چلایا تے کدو بیج دتے۔ اوہناں کھیت اندر کھاد وی پائی تے پانی وی لایا۔ کجھ دناں پچھوں کھیت وچ ہری ہری گھاہ اگ آئی۔
اوہناں دے کھیت دے نال ای اک وڈے زمین دار دا ڈیرہ سی۔ اک دن اوس دے گھوڑے کھل کے گھاہ چرن لئی اوہناں دے کھیت وچ آ گئے۔ اوہناں چاراں جدوں گھوڑے ویکھے تاں آکھیا کہ ساڈے کھیت وچ گھوڑے اگ آئے نیں۔ اوہناںگھوڑیاں نوں پھڑ کے بنھ لیا۔ شام نوں زمین دار نوں پتہ لگا کہ اوہدے گھوڑے گوانڈھیاں بنھ لے نیں۔ اوہ گھوڑے لین لئی اوہناں کول گیا۔ اوہناں چاراں آکھیا کہ اساں اپنے کھیت وچ گھوڑے دے انڈے بیجے سن۔ ایہہ گھوڑے اگے نیں۔ ایس لئی ایہہ ساڈے گھوڑے نیں۔ زمین دار نوں پتہ لگا کہ ایہہ چاروں پاگل نیں۔ اوہ ایہناں نوں نال لے گیا تے اوہناں نوں ملازم رکھ لیا۔
زمین دار نے پہلے نوں آکھیا کہ توں ہل واہ کے پکان لئی گوشت لے کے آنا ایں۔ دوجے نوں آکھیا کہ توں بھیڈ بکریاں لے کے جانیاں ایں۔ تیجے نوں آکھیا کہ توں ریڑھی تے توڑی لے کے آنی ایں۔ چوتھے نوں آکھیا کہ جدوں میری ماں سوں جاوے توں اوہدے منہ توں مکھیاں اڈانیاں نیں۔ پہلا ہل لے کے کھیت چلا جاندا اے۔ ہل واہن توں بعد اوس اک...
Technological advancement makes translation convenient due to the emergence of various translation tools. This Explanatory-Sequential study aims to determine the preference and the factors affecting the preference of Filipino and Foreign college students toward the Online Translation Tool. Likewise, it also aimed to identify if there is a significant difference between the respondents' choices. To acquire the data, the researchers used a survey conducted on 15 Filipino and foreign collegiate students enrolled in universities in Manila and a focus group discussion among 3 Filipino students. The transcribed data were analyzed using Thematic Analysis. Moreover, the results of the quantitative data revealed that Google Translate was the preferred Online Translation Tool of Filipino and Foreign students for the reason of accessibility, user-friendliness, and the tendency of users to brand bias. The reasons presented in quantitative data are strengthened by the themes identified in the qualitative data. The three prevailing themes of Brand Bias, Accessibility, and Accuracy constituted the central theme of User Friendliness. It was identified through the Chi-Square Test that there is no significant difference among respondents' preferences (0.345 P-Value) toward online Translation software.
Automobiles release a number of toxic metals into the surrounding environment. They enter human body through food chain and cause many toxic effects. Plants prove good indicators of their existence. In this study five herbaceous plant species (Calotropis procera, Datura alba, Ricinus communis, Parthenium hysterophorus and Cenchrus ciliaris) commonly growing along two roads i.e. Motorway (M-2) and Faisalabad-Sargodha road (FSR) in the Punjab, Pakistan, were collected. Plant and soil samples were collected in all the four seasons (2013-2014) from roadsides. Samples taken 100 m away from roads were designated as control. Lead (Pb), cadmium (Cd), nickel (Ni) and zinc (Zn) metals were analyzed in all the plants and soil samples by ICP-AES (Inductively Coupled Plasma Atomic Emission Spectrophotometer). Carbon (C) and nitrogen (N) contents in plants and soil samples were also measured. Relative plant attributes i.e. photosynthetic pigments (chlorophyll a, b, total chlorophyll contents and carotenoids), gas exchange characters (photosynthetic rate (A), transpiration rate (E), stomatal conductance (gs), internal CO2 concentration (Ci) and water use efficiency (WUE)), total soluble proteins, total free amino acids and total antioxidant activity were studied. Significantly higher concentrations of all the metals were found along roadsides in plants and soils as compared to controls and they clearly showed spatial and temporal variations. In both plants and soil samples, higher contamination of metals was recorded during summer season, while, the least contamination was noticed during winter season. The metals concentrations were obtained in the order Zn > Ni > Pb > Cd. Higher metals, C and N concentrations were recorded along FSR road as compared to M-2. Pull-111 was proved to be the most polluted site. A general reduction in photosynthetic pigments, photosynthetic rate, transpiration rate, stomatal conductance and total soluble proteins were recorded, whereas, increase in internal CO2 concentration, water use efficiency, total free amino acids and total antioxidant activity was observed under metal toxicity. Among plants, Calotropis procera leaves accumulated the highest level of Pb, Cd and Ni, while, Ricinus communis showed tendency to accumulate high quantities of Zn, thus, these plant species can be used as good biomonitors / phytoremediators. The metal contents in plants at most of the sites showed significant positive correlation with traffic density. High level of metals was also found in fuel (petrol and diesel) and soot samples. So, control measures are required to overcome transport sector related pollution which may become severe in forthcoming days.