Applying Word2vec on a dataset of texts

AC24

Hi, I created ad SVM model for text classification and I want to share the process with you to have any advise to improve it. The purpose of this classificator is to classify a dataset of comments, reviews, or sentences in general, into positive and negative and the dataset I used for its training was made of 2400 tweets (1200 positive and 1200 negative).

我也会问你如果有一个implemen方式t in this process a word2vec embedding, or how can I create an alternative process for this purpose. If I try to apply word2vec on the opetators loop and loop collection, this return some errors and I don't know how to give as input a dataset of sentences (my dataset has the attributs "text" and "sentiment") and not a whole text or a collaction of long texts.

Here is the code of the process:

<运营商激活= " true " class = "process" compatibility="9.10.011" expanded="true" name="Process">

<参数键= " logverbosity " value = " init " / >

<帕拉meter key="random_seed" value="2001"/>

<帕拉meter key="send_mail" value="never"/>

<帕拉meter key="notification_email" value=""/>

<帕拉meter key="process_duration_for_mail" value="30"/>

<帕拉meter key="encoding" value="SYSTEM"/>

<运营商激活= " true " class = "retrieve" compatibility="9.10.011" expanded="true" height="68" name="Retrieve clean_dataset2" width="90" x="45" y="136">

<帕拉meter key="repository_entry" value="clean_dataset2"/>

<运营商激活= " true " class = "operator_toolbox:extract_sentiment" compatibility="2.14.000" expanded="true" height="103" name="Extract Sentiment" width="90" x="179" y="136">

<帕拉meter key="model" value="vader"/>

<帕拉meter key="text_attribute" value="text"/>

<帕拉meter key="show_advanced_output" value="true"/>

<帕拉meter key="use_default_tokenization_regex" value="true"/>

<帕拉meter key="tokenization_regex" value=".*"/>

<运营商激活= " true " class = "nominal_to_text" compatibility="9.10.011" expanded="true" height="82" name="Nominal to Text" width="90" x="313" y="136">

<帕拉meter key="attribute_filter_type" value="single"/>

<帕拉meter key="attribute" value="text"/>

<帕拉meter key="attributes" value="text|sentiment"/>

<帕拉meter key="use_except_expression" value="false"/>

<帕拉meter key="value_type" value="nominal"/>

<帕拉meter key="use_value_type_exception" value="false"/>

<帕拉meter key="except_value_type" value="file_path"/>

<帕拉meter key="block_type" value="single_value"/>

<帕拉meter key="use_block_type_exception" value="false"/>

<帕拉meter key="except_block_type" value="single_value"/>

<帕拉meter key="invert_selection" value="false"/>

<帕拉meter key="include_special_attributes" value="false"/>

<运营商激活= " true " class = "text:process_document_from_data" compatibility="9.4.000" expanded="true" height="82" name="Process Documents from Data" width="90" x="447" y="136">

<帕拉meter key="create_word_vector" value="true"/>

<帕拉meter key="vector_creation" value="TF-IDF"/>

<帕拉meter key="add_meta_information" value="true"/>

<帕拉meter key="keep_text" value="true"/>

<帕拉meter key="prune_method" value="percentual"/>

<帕拉meter key="prune_below_percent" value="1.0"/>

<帕拉meter key="prune_above_percent" value="99.0"/>

<帕拉meter key="prune_below_rank" value="0.05"/>

<帕拉meter key="prune_above_rank" value="0.95"/>

<帕拉meter key="datamanagement" value="double_sparse_array"/>

<帕拉meter key="data_management" value="auto"/>

<帕拉meter key="select_attributes_and_weights" value="false"/>

<运营商激活= " true " class = "text:tokenize" compatibility="9.4.000" expanded="true" height="68" name="Tokenize" width="90" x="179" y="34">

<帕拉meter key="mode" value="non letters"/>

<帕拉meter key="characters" value=".:"/>

<帕拉meter key="language" value="English"/>

<帕拉meter key="max_token_length" value="3"/>

<运营商激活= " true " class = "text:transform_cases" compatibility="9.4.000" expanded="true" height="68" name="Transform Cases" width="90" x="313" y="34">

<帕拉meter key="transform_to" value="lower case"/>

<运营商激活= " true " class = "text:filter_stopwords_english" compatibility="9.4.000" expanded="true" height="68" name="Filter Stopwords (English)" width="90" x="447" y="34"/>

<运营商激活= " true " class = "text:filter_by_length" compatibility="9.4.000" expanded="true" height="68" name="Filter Tokens (by Length)" width="90" x="581" y="34">

<帕拉meter key="min_chars" value="4"/>

<帕拉meter key="max_chars" value="999"/>

<运营商激活= " true " class = "select_attributes" compatibility="9.10.011" expanded="true" height="82" name="Select Attributes" width="90" x="581" y="136">

<帕拉meter key="attribute_filter_type" value="no_missing_values"/>

<帕拉meter key="attribute" value=""/>

<帕拉meter key="attributes" value="|sentiment|text"/>

<帕拉meter key="use_except_expression" value="false"/>

<帕拉meter key="value_type" value="attribute_value"/>

<帕拉meter key="use_value_type_exception" value="false"/>

<帕拉meter key="except_value_type" value="time"/>

<帕拉meter key="block_type" value="attribute_block"/>

<帕拉meter key="use_block_type_exception" value="false"/>

<帕拉meter key="except_block_type" value="value_matrix_row_start"/>

<帕拉meter key="invert_selection" value="false"/>

<帕拉meter key="include_special_attributes" value="true"/>

<运营商激活= " true " class = "store" compatibility="9.10.011" expanded="true" height="68" name="Store wordlist" width="90" x="380" y="289">

<帕拉meter key="repository_entry" value="Wordlist"/>

<运营商激活= " true " class = "set_role" compatibility="9.10.011" expanded="true" height="82" name="Set Role" width="90" x="715" y="136">

<帕拉meter key="attribute_name" value="sentiment"/>

<帕拉meter key="target_role" value="label"/>

<帕拉meter key="Score" value="regular"/>

<运营商激活= " true " class = "split_data" compatibility="9.10.011" expanded="true" height="103" name="Split Data" width="90" x="581" y="238">

<帕拉meter key="ratio" value="0.7"/>

<帕拉meter key="ratio" value="0.3"/>

<帕拉meter key="sampling_type" value="stratified sampling"/>

<帕拉meter key="use_local_random_seed" value="false"/>

<帕拉meter key="local_random_seed" value="1992"/>

<运营商激活= " true " class = "concurrency:cross_validation" compatibility="9.10.011" expanded="true" height="145" name="Cross Validation" width="90" x="715" y="238">

<帕拉meter key="split_on_batch_attribute" value="false"/>

<帕拉meter key="leave_one_out" value="false"/>

<帕拉meter key="number_of_folds" value="10"/>

<帕拉meter key="sampling_type" value="automatic"/>

<帕拉meter key="use_local_random_seed" value="false"/>

<帕拉meter key="local_random_seed" value="1992"/>

<帕拉meter key="enable_parallel_execution" value="true"/>

<运营商激活= " true " class = "support_vector_machine_libsvm" compatibility="9.10.011" expanded="true" height="82" name="SVM (2)" width="90" x="179" y="136">

<帕拉meter key="svm_type" value="C-SVC"/>

<帕拉meter key="kernel_type" value="linear"/>

<帕拉meter key="degree" value="3"/>

<帕拉meter key="gamma" value="0.05"/>

<帕拉meter key="coef0" value="0.0"/>

<帕拉meter key="C" value="0.0"/>

<帕拉meter key="nu" value="0.5"/>

<帕拉meter key="cache_size" value="80"/>

<帕拉meter key="epsilon" value="1.0"/>

<帕拉meter key="p" value="0.1"/>

<帕拉meter key="shrinking" value="true"/>

<帕拉meter key="calculate_confidences" value="true"/>

<帕拉meter key="confidence_for_multiclass" value="false"/>

<运营商激活= " true " class = "apply_model" compatibility="9.10.011" expanded="true" height="82" name="Apply Model" width="90" x="112" y="85">

<帕拉meter key="create_view" value="false"/>

<运营商激活= " true " class = "performance" compatibility="9.10.011" expanded="true" height="82" name="Performance" width="90" x="179" y="187">

<帕拉meter key="use_example_weights" value="true"/>

<运营商激活= " true " class = "store" compatibility="9.10.011" expanded="true" height="68" name="Store model" width="90" x="849" y="238">

<帕拉meter key="repository_entry" value="My SVM model"/>

<运营商激活= " true " class = "apply_model" compatibility="9.10.011" expanded="true" height="82" name="Apply Model (2)" width="90" x="447" y="442">

<帕拉meter key="create_view" value="false"/>

<运营商激活= " true " class = "performance" compatibility="9.10.011" expanded="true" height="82" name="Performance (2)" width="90" x="648" y="442">

<帕拉meter key="use_example_weights" value="true"/>

I also share an image of the process.