test
And some ML methods were attempted in previous studies, like Random forests methods with land use information over USA by Hu (2017) (CV R2 =0.80) and over china by G. Chen (2018) (CV R2 =0.83, RMSE =28.10μg⁄m^3 ); artificial neural network (ANN) in east coast of the United States by Gupta and Christopher (2009) (R =0.78); K-nearest neighbors (CV R2 =0.62), Decision tree (CV R2 =0.34), Support vector machine (CV R2 =0.61), Random forests (CV R2 =0.74), Gradient boost model (CV R2 =0.79) with land use data in North China reported by Jiang and Christakos (2017); Bagged tree (CV R2 =0.68), Support vector machine (CV R2 =0.61), Random forests (CV R2 =0.79), Elastic net regression (CV R2 =0.54), Gradient boost model (CV R2 =0.79) with Chemical transport models and land use data in Northern California reported by Reid (2015). Although some land use data used in previous studies like urban cover, population density and point emissions were not included in our model, our combined method still has high predictive ability (CV R2 =0.86) to estimate PM2.5 level in more areas (86.17% coverage), with considering potential lag effects of AOD and meteorological factors, capturing complicated spatial or temporal heterogeneity and restricting the overfit by regularized boosting. Compared with the ML techniques or combined method tested in our study (shown in Table A.10), the combine method provides a more robust, effective and computation-time saving way to estimate PM2.5 concentrations, with the help of regularized boosting, lagged model and parallel processing.
Transition Words or Phrases used:
if, so, still
Attributes: Values AverageValues Percentages(Values/AverageValues)% => Comments
Performance on Part of Speech:
To be verbs : 2.0 13.1623246493 15% => More to be verbs wanted.
Auxiliary verbs: 0.0 7.85571142285 0% => OK
Conjunction : 12.0 10.4138276553 115% => OK
Relative clauses : 0.0 7.30460921844 0% => More relative clauses wanted.
Pronoun: 3.0 24.0651302605 12% => OK
Preposition: 31.0 41.998997996 74% => OK
Nominalization: 3.0 8.3376753507 36% => More nominalizations (nouns with a suffix like: tion ment ence ance) wanted.
Performance on vocabulary words:
No of characters: 1282.0 1615.20841683 79% => OK
No of words: 247.0 315.596192385 78% => More content wanted.
Chars per words: 5.19028340081 5.12529762239 101% => OK
Fourth root words length: 3.96437052324 4.20363070211 94% => OK
Word Length SD: 2.88411045769 2.80592935109 103% => OK
Unique words: 149.0 176.041082164 85% => More unique words wanted.
Unique words percentage: 0.603238866397 0.561755894193 107% => OK
syllable_count: 345.6 506.74238477 68% => OK
avg_syllables_per_word: 1.4 1.60771543086 87% => OK
A sentence (or a clause, phrase) starts by:
Pronoun: 1.0 5.43587174349 18% => OK
Article: 1.0 2.52805611222 40% => OK
Subordination: 1.0 2.10420841683 48% => OK
Conjunction: 1.0 0.809619238477 124% => OK
Preposition: 3.0 4.76152304609 63% => OK
Performance on sentences:
How many sentences: 4.0 16.0721442886 25% => Need more sentences. Double check the format of sentences, make sure there is a space between two sentences, or have enough periods. And also check the lengths of sentences, maybe they are too long.
Sentence length: 61.0 20.2975951904 301% => The Avg. Sentence Length is relatively long.
Sentence length SD: 175.992187327 49.4020404114 356% => The lengths of sentences changed so frequently.
Chars per sentence: 320.5 106.682146367 300% => Less chars_per_sentence wanted.
Words per sentence: 61.75 20.7667163134 297% => Less words per sentence wanted.
Discourse Markers: 3.25 7.06120827912 46% => More transition words/phrases wanted.
Paragraphs: 1.0 4.38176352705 23% => More paragraphs wanted.
Language errors: 0.0 5.01903807615 0% => OK
Sentences with positive sentiment : 4.0 8.67935871743 46% => More positive sentences wanted.
Sentences with negative sentiment : 0.0 3.9879759519 0% => More negative sentences wanted.
Sentences with neutral sentiment: 0.0 3.4128256513 0% => More facts, knowledge or examples wanted.
What are sentences with positive/Negative/neutral sentiment?
Coherence and Cohesion:
Essay topic to essay body coherence: 0.0267117179761 0.244688304435 11% => The similarity between the topic and the content is low.
Sentence topic coherence: 0.0286651594095 0.084324248473 34% => Sentence topic similarity is low.
Sentence topic coherence SD: 0.0496495125044 0.0667982634062 74% => OK
Paragraph topic coherence: 0.0267117179761 0.151304729494 18% => Maybe some paragraphs are off the topic.
Paragraph topic coherence SD: 0.0 0.056905535591 0% => Paragraphs are similar to each other. Some content may get duplicated or it is not exactly right on the topic.
Essay readability:
automated_readability_index: 33.9 13.0946893788 259% => Automated_readability_index is high.
flesch_reading_ease: 26.48 50.2224549098 53% => Flesch_reading_ease is low.
smog_index: 13.0 7.44779559118 175% => OK
flesch_kincaid_grade: 24.7 11.3001002004 219% => Flesch kincaid grade is high.
coleman_liau_index: 13.71 12.4159519038 110% => OK
dale_chall_readability_score: 11.58 8.58950901804 135% => OK
difficult_words: 77.0 78.4519038076 98% => OK
linsear_write_formula: 26.5 9.78957915832 271% => Linsear_write_formula is high.
gunning_fog: 26.4 10.1190380762 261% => Gunning_fog is high.
text_standard: 27.0 10.7795591182 250% => The average readability is very high. Good job!
What are above readability scores?
---------------------
Minimum 250 words wanted.
Minimum four paragraphs wanted.
It is not exactly right on the topic in the view of e-grader. Maybe there is a wrong essay topic.
Rates: 11.2359550562 out of 100
Scores by essay e-grader: 0.67 Out of 6
---------------------
Note: the e-grader does NOT examine the meaning of words and ideas. VIP users will receive further evaluations by advanced module of e-grader and human graders.