Journal of applied research in chemisry
,
Issue4,Year,
Autumn
1389
براساس اهمیت پیش بینی نقطه های ذوب ترکیب ها، در این مقاله سعی شد که برای دسته وسیعی از ترکیب ها مدل مناسبی که توانایی پیش بینی نقطه های ذوب را در حد مطلوبی داشته باشد، ارایه شود. برای این منظور 4173 ترکیب شیمیایی با ساختارهای متنوع گزارش شده در مقاله های قبلی، انتخاب More
براساس اهمیت پیش بینی نقطه های ذوب ترکیب ها، در این مقاله سعی شد که برای دسته وسیعی از ترکیب ها مدل مناسبی که توانایی پیش بینی نقطه های ذوب را در حد مطلوبی داشته باشد، ارایه شود. برای این منظور 4173 ترکیب شیمیایی با ساختارهای متنوع گزارش شده در مقاله های قبلی، انتخاب و برای توصیف ساختار آن ها از یک دسته 202 تایی از توصیفکننده های 2D و 3D استفاده شد. این دسته داده ها به دو دسته آموزش و دسته تست به ترتیب با اندازه های 2573 و 1600 تقسیم بندی شد. ارزیابی بیشتر مدل ایجاد شده به کمک یک دسته داده دیگر با اندازه 277 صورت گرفت. برای کاهش حجم ماتریس توصیف کننده ها از تجزیه اجزای اصلی کمک گرفته شد و از شبکه عصبی برای ایجاد مدل استفاده شد. بردارهای ویژه به دست آمده از تجزیه اجزای اصلی بر اساس معیارهای متفاوتی مرتب و به عنوان ورودی شبکه مورد استفاده قرار گرفتند. معیارهای استفاده شده عبارت بودند از بزرگی مقدار ویژه، میزان همبستگی با نقطه ذوب و میزان قدرت پیش بینی کنندگی. بهترین مدل در حالتی بدست آمد که بردارهای ویژه براساس میزان قدرت پیش بینی کنندگی اشان مرتب و به عنوان ورودی استفاده بودند. در پایان پارامترهای شبکه از قبیل تعداد لایه های پنهان، تعداد گره در هر لایه، سرعت یادگیری و مومنتوم بهینه سازی شدند. شبکه با ساختار [25 10 8 1] و سرعت یادگیری 0/7 و مومنتوم 0/16به عنوان شبکه بهینه انتخاب شد.
Manuscript profile
Journal of the Iranian Chemical Research
,
Issue2,Year,
Spring
2011
The classification and regression trees (CART) possess the advantage of being able to handlelarge data sets and yield readily interpretable models. In spite to these advantages, they are alsorecognized as highly unstable classifiers with respect to minor perturbations i More
The classification and regression trees (CART) possess the advantage of being able to handlelarge data sets and yield readily interpretable models. In spite to these advantages, they are alsorecognized as highly unstable classifiers with respect to minor perturbations in the training data.In the other words methods present high variance. Fuzzy logic brings in an improvement in theseaspects due to the elasticity of fuzzy sets formalism. ACS, which is a meta-heuristic algorithmand derived from the observation of real ants, was used to optimize fuzzy parameters. Thepurpose of this study was to explore the use of fuzzy regression tree (RT) for modeling ofmelting points of a large variety of chemical compounds. To test the ability of the resulted tree, aset of approximately 4173 structures and their melting points were used (3000 compounds astraining set and 1173 as validation set). Further, an external test set contains of 277 drugs wereused to validate the prediction ability of the tree. Comparison the results obtained from both treesshowed that the fuzzy RT performs better than that produced by recursive partitioning procedure.
Manuscript profile
Journal of the Iranian Chemical Research
,
Issue4,Year,
Summer
2012
Toxicity of 38 aliphatic carboxylic acids was studied using non-linear quantitative structure-toxicityrelationship (QSTR) models. The adaptive neuro-fuzzy inference system (ANFIS) was used to construct thenonlinear QSTR models in all stages of study. Two ANFIS models we More
Toxicity of 38 aliphatic carboxylic acids was studied using non-linear quantitative structure-toxicityrelationship (QSTR) models. The adaptive neuro-fuzzy inference system (ANFIS) was used to construct thenonlinear QSTR models in all stages of study. Two ANFIS models were developed based upon differentsubsets of descriptors. The first one used log ow K and LUMO E as inputs and had good prediction ability; forthe training set of 28 compounds 2Training R was 0.86 and for the test set of 10 compounds, the correspondingstatistic was 2Test R =0.97. Two outliers were detected for this ANFIS model and removing them improved thequality of the model. Another ANFIS model was constructed based on PEOE_VSA_FPNEG and G3udescriptors chosen by exhaustive search of all two combinations of calculated descriptors by Dragon andMOE softwares. The later ANFIS model showed better performance than the former ( 2Training R =0.92 and2Test R =0.90) and no outlier was detected.
Manuscript profile
Journal of the Iranian Chemical Research
,
Issue5,Year,
Autumn
2012
In this study, aluminosilicate MCM-41 has been found to be an efficient catalyst for reaction betweendifferent benzaldehyde derivatives and 5,5-dimethyl-1,3-cyclohexanedione to give open chain analogue of1,8-dioxo-octahydroxanthene in excellent yields. No evidence for t More
In this study, aluminosilicate MCM-41 has been found to be an efficient catalyst for reaction betweendifferent benzaldehyde derivatives and 5,5-dimethyl-1,3-cyclohexanedione to give open chain analogue of1,8-dioxo-octahydroxanthene in excellent yields. No evidence for the formation of cyclization products wasobserved. The catalyst can be used several times after recovery.
Manuscript profile
Sanad
Sanad is a platform for managing Azad University publications